WO2022048193A1 - Map drawing method and apparatus - Google Patents
Map drawing method and apparatus Download PDFInfo
- Publication number
- WO2022048193A1 WO2022048193A1 PCT/CN2021/094917 CN2021094917W WO2022048193A1 WO 2022048193 A1 WO2022048193 A1 WO 2022048193A1 CN 2021094917 W CN2021094917 W CN 2021094917W WO 2022048193 A1 WO2022048193 A1 WO 2022048193A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- picture
- positioning
- pictures
- similarity
- road
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
- G09B29/003—Maps
- G09B29/005—Map projections or methods associated specifically therewith
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
Definitions
- the embodiments of the present application relate to the field of image processing, and in particular, to a map drawing method and device.
- high-precision positioning maps that provide positioning information for autonomous driving equipment have also received extensive attention.
- high-precision positioning maps can provide more detailed map data, and can be directly identified and used by devices (such as vehicles for autonomous driving), thereby providing accurate positioning information for autonomous driving of vehicles. It can be understood that the higher the accuracy of the high-precision positioning map, the more accurate the positioning information can be provided for the vehicle, which is more conducive to the automatic driving of the vehicle.
- the current high-precision map cannot provide accurate height information, which makes it impossible for vehicles to distinguish between roads with different road distributions at different heights (such as interchanges in cities, etc.) according to the current high-precision map. This also makes the automatic driving of the vehicle prone to problems, such as the vehicle cannot achieve precise positioning in a complex environment.
- the embodiments of the present application provide a map drawing method and device, which solve the problem that the existing positioning map cannot provide accurate positioning information.
- a first aspect provides a map drawing method, the method comprising: acquiring laser point cloud data of a first area to be drawn, where the first area to be drawn includes a first road and a second road on different planes; according to the laser point cloud data , obtain the set of the first positioning picture and the set of the second positioning picture; wherein, the reference heights of any two first positioning pictures in each set (such as the set of the first positioning picture, or the set of the second positioning picture) are the same Exceed the first threshold; the reference height difference of any two second positioning pictures does not exceed the first threshold; the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is the corresponding height of the collected positioning picture.
- the laser point cloud data When the laser point cloud data is obtained, it corresponds to the height of the road where the device is located; the first threshold and the second threshold are both positive numbers; the positioning pictures in the first positioning picture set are fused to obtain the map corresponding to the first road, and the second The positioning pictures in the positioning picture set are fused to obtain a map corresponding to the second road.
- a solution that can solve the problem that the positioning information at different heights cannot be accurately judged.
- a corresponding positioning map may be drawn according to roads at different heights, so as to obtain positioning information by referring to the positioning map under the corresponding road layer according to the road where the vehicle is currently located.
- sets of positioning pictures of the first road and the second road in different road layers can be obtained respectively, and the set of positioning pictures corresponding to each road layer is the positioning map corresponding to the road layer.
- different road layers can be distinguished according to the absolute height of the road surface under the global coordinate system (that is, the reference height of the corresponding positioning picture) where the corresponding device (such as the vehicle that collects laser point cloud data) is located when the road layer is obtained.
- the positioning pictures corresponding to the different laser point cloud data are in the same road layer, that is, the different laser point clouds
- the positioning pictures corresponding to the data are in the same positioning picture set.
- the first threshold may be the same as the second threshold, and the second threshold may also be greater than the first threshold.
- the setting of the threshold can be flexibly selected according to the actual situation. In this way, in different road scenarios, especially in road scenarios with complex height distribution, accurate positioning information can be provided for devices that need to be positioned, such as vehicles.
- the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; Before the first positioning picture set and the second positioning picture set, the method further includes: according to the three-dimensional coordinate information of the first position, obtaining relative height information of the first position relative to the road on which the laser point cloud data is collected. Based on this solution, the height information of different positions (eg, the first position) in the scene is identified by the relative height information. In this way, under the premise of compliance, the height of objects in the scene can be marked. Exemplarily, according to the absolute height information in the laser point cloud data, combined with the absolute height of the road surface in the form of the vehicle when the laser point cloud data is collected, the corresponding relative height information can be determined.
- the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane. Based on this solution, it is determined whether the first position is a lane mark through the first mark, so that more accurate road surface information can be provided accordingly in the subsequently determined positioning map, for example, the position of the lane mark on the current road surface.
- obtaining the first positioning picture set and the second positioning picture set according to the laser point cloud data includes: obtaining a plurality of positioning pictures according to the laser point cloud data, and the laser beam corresponding to each positioning picture
- the collection time of the point cloud data is within the preset range, and the pixel value of the pixel included in each positioning picture is determined by the relative height information of the corresponding position of the pixel and the first identification of the laser point cloud data corresponding to the position;
- the positioning pictures in the same road layer are fused to obtain the first positioning picture set and the second positioning picture set.
- the laser point cloud data with the same horizontal coordinates can be compressed into pixel points with different pixel values, so as to obtain the two-dimensional distribution that can characterize the object in the horizontal direction. location map.
- the pixel value such as gray value
- the positioning picture obtained according to this example It is also possible to restore the three-dimensional distribution of objects in the corresponding area after certain processing. In this way, in the subsequent use of the positioning picture, by controlling the above processing method of restoring the three-dimensional distribution, more detailed positioning information can be provided in different regions under the premise of compliance.
- acquiring the first positioning picture set and the second positioning picture set includes: determining the similarity between the first positioning picture and the second positioning picture, and determining the first positioning picture and the second positioning picture according to the similarity. Whether the two positioning pictures are in the same road layer, the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture; wherein, the first positioning picture and the second positioning picture are multiple positioning pictures, any two have The positioning pictures of the same horizontal coverage area; the positioning pictures in the same road layer among the multiple positioning pictures are fused to obtain the first positioning picture set corresponding to the first road and the second positioning picture corresponding to the second road gather. Based on this solution, a possible solution for determining positioning pictures in the same road layer is provided.
- the similarity between the first picture and the second picture with the same horizontal coverage area can be determined, and whether the two positioning pictures are in the same position can be determined according to the similarity.
- the location image in the roads layer Exemplarily, when the similarity is higher than the corresponding preset threshold, it is determined that the two positioning pictures are positioning pictures in the same road layer.
- the similarity is lower than the corresponding preset advance, it is determined that the two positioning pictures are positioning pictures in different road layers.
- the determining the similarity between the first positioning picture and the second positioning picture includes: determining the similarity between the first positioning picture and the second positioning picture according to local features of the first positioning picture and the second positioning picture
- the first similarity; the local features include one or more of the following: the average value of the grayscale of the pixels in the positioning picture, the grayscale variance of the pixels, and the grayscale covariance of the pixels; according to the similarity, determine the first positioning picture and Whether the second positioning picture is in the same road layer includes: when the first similarity is greater than the first threshold, the first positioning picture and the second positioning picture are in the same road layer.
- a possible solution for determining the similarity is provided, that is, determining the similarity between two positioning pictures based on the comparison of local features.
- the difference of local features can be accurately evaluated, and then the corresponding similarity can be obtained. Therefore, a relatively accurate similarity measurement can be effectively performed on the positioning pictures with relatively simple scenes included in the positioning pictures.
- the method further includes: when the first similarity is less than the first threshold, the first positioning picture and the second positioning picture are in different road layers. Based on this solution, a solution for determining that the two positioning images are not in the same road layer is provided. That is, according to the magnitude relationship between the first similarity and the first threshold, it is determined that the first positioning picture and the second positioning picture are not in the same road layer.
- determining the similarity between the first positioning picture and the second positioning picture includes: determining the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture.
- the second similarity of the two positioning pictures according to the similarity, determining whether the first positioning picture and the second positioning picture are in the same road layer, including: when the second similarity is less than the second threshold, the first positioning picture and the second positioning picture are in the same road layer.
- the positioning image is in the same road layer.
- the similarity can be determined according to the relative height information corresponding to different pixel points. It should be understood that the pixel values of different pixels may be determined according to the relative height information of the object corresponding to the horizontal position. Therefore, in some implementations, the first positioning picture and The similarity of the second positioning picture.
- the method further includes: when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers. Based on this solution, another solution for determining that the two positioning pictures are not in the same road layer is provided.
- determining the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture including: for the first positioning picture Perform the following operations respectively with the second positioning picture to obtain the feature fingerprint corresponding to the first positioning picture, and the feature fingerprint corresponding to the second positioning picture: delete the pixels of the preset row and/or preset column in the positioning picture to obtain the reduced According to the average value of the relative height of each pixel in the reduced positioning picture, normalize the reduced positioning picture, and determine the corresponding positioning picture according to the pixel values of the reduced positioning picture after normalization.
- the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture determine the second similarity of the first positioning picture and the second positioning picture, and the second similarity is the feature fingerprint of the first positioning picture and the second similarity.
- the Hamming distance of the feature fingerprint of the second positioning image Based on this scheme, a possible implementation of determining similarity based on relative height information is clarified. It can be seen that the feature fingerprint in this example is the global information that can more accurately reflect the positioning picture. Therefore, it can be determined whether the two positioning maps are in the same road layer through the similarity in the global state. In some implementation scenarios, this solution can better measure the similarity of complex positioning pictures.
- a map drawing device comprising: an acquisition unit and a fusion unit.
- the acquisition unit is used to acquire the laser point cloud data of the first area to be drawn, and the first area to be drawn includes the first road and the second road on different planes; the acquisition unit is also used to acquire the laser point cloud data according to the laser point cloud data.
- the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; The three-dimensional coordinate information of the first position is obtained, and the relative height information of the first position relative to the road on which the laser point cloud data is collected is obtained.
- the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
- the acquisition unit is configured to acquire multiple positioning pictures according to the laser point cloud data, the acquisition time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture
- the pixel value of the pixel included in the pixel value is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position;
- the fusion unit is used to combine the positioning pictures in the same road layer among the multiple positioning pictures Perform fusion to obtain the first positioning picture set and the second positioning picture set.
- the apparatus further includes: a determining unit, where the determining unit is configured to determine the similarity between the first positioning picture and the second positioning picture.
- the determining unit is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, and the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture; wherein the first positioning picture and the second positioning picture are similar.
- the positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures.
- the fusion unit is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
- the apparatus further includes: a determining unit, configured to determine the first similarity between the first positioning picture and the second positioning picture according to the local features of the first positioning picture and the second positioning picture ;
- the local features include one or more of the following: the grayscale average value of the pixels in the positioning picture, the grayscale variance of the pixels, and the grayscale covariance of the pixels; the determining unit is also used for when the first similarity is greater than the first similarity.
- a threshold is used, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
- the determining unit is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
- the determining unit is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture; When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
- the determining unit is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
- the acquisition unit is specifically configured to perform the following operations on the first positioning picture and the second positioning picture respectively, to obtain the feature fingerprint corresponding to the first positioning picture and the feature fingerprint corresponding to the second positioning picture: Delete the pixels in the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the average value of the relative heights of each pixel in the reduced positioning picture.
- Each pixel value of the reduced positioning picture after normalization is determined to determine the feature fingerprint corresponding to the positioning picture; according to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, the first positioning picture and the second positioning picture are determined.
- the second similarity is the Hamming distance between the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture.
- a mapping device comprising one or more processors and one or more memories; one or more memories coupled to the one or more processors, the one or more memories A memory stores computer instructions; when executed by the one or more processors, the computer instructions are caused to cause the communication device to perform the mapping method of any one of the first aspect and possible designs thereof.
- the processor when invoking the computer instructions in the memory, is configured to acquire the laser point cloud data of the first area to be drawn, where the first area to be drawn includes the first road and the second road on different planes; also used for According to the laser point cloud data, a set of first positioning pictures and a set of second positioning pictures are obtained; wherein, the reference height difference of any two first positioning pictures does not exceed the first threshold; the reference heights of any two second positioning pictures The difference does not exceed the first threshold; the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is the height of the road corresponding to the device when the laser point cloud data corresponding to the positioning picture is collected; The first threshold and the second threshold are both positive numbers; the processor is configured to fuse the positioning pictures in the first positioning picture set, obtain a map corresponding to the first road, and fuse the positioning pictures in the second positioning picture set, Obtain the map corresponding to the second road.
- the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; the processor is further configured to The three-dimensional coordinate information of the first position is obtained, and the relative height information of the first position relative to the road on which the laser point cloud data is collected is obtained.
- the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
- the processor is used to acquire multiple positioning pictures according to the laser point cloud data, the collection time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture
- the pixel value of the pixel included in the pixel value is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position;
- the processor is used for positioning the positioning pictures in the same road layer among the multiple positioning pictures Perform fusion to obtain the first positioning picture set and the second positioning picture set.
- the apparatus further includes: a processor configured to determine the similarity between the first positioning picture and the second positioning picture.
- the processor is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, and the similarity is used to indicate the similarity of the first positioning picture and the second positioning picture; wherein, the first positioning picture and the second positioning picture are similar.
- the positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures.
- the processor is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
- the apparatus further includes: a processor, configured to determine the first similarity between the first positioning picture and the second positioning picture according to local features of the first positioning picture and the second positioning picture;
- the local features include one or more of the following: the grayscale average value of the pixels in the positioning image, the grayscale variance of the pixels, and the grayscale covariance of the pixels; the processor is also used for when the first similarity is greater than the first similarity When the threshold is set, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
- the processor is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
- the processor is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture; When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
- the processor is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
- the processor is specifically configured to perform the following operations on the first positioning picture and the second positioning picture, respectively, to obtain a feature fingerprint corresponding to the first positioning picture and a feature fingerprint corresponding to the second positioning picture: Delete the pixels in the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the average value of the relative heights of each pixel in the reduced positioning picture.
- Each pixel value of the reduced positioning picture after normalization is determined to determine the feature fingerprint corresponding to the positioning picture; according to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, the first positioning picture and the second positioning picture are determined.
- the second similarity is the Hamming distance between the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture.
- a chip system in a fourth aspect, is provided, and the chip system can be applied to a map drawing device.
- the chip system includes an interface circuit and a processor; the interface circuit and the processor are interconnected by a line; the interface circuit is used to receive a signal from a memory of the electronic device and send a signal to the processor, and the signal includes computer instructions stored in the memory. ;
- the processor executes the computer instructions, the chip system executes the map drawing method as described in any one of the first aspect and its possible designs.
- a computer-readable storage medium includes computer instructions that, when executed, perform the mapping method described in any one of the first aspect and possible designs thereof. .
- a sixth aspect provides a computer program product, the computer program product includes instructions, when the computer program product runs on a computer, the computer can execute any one of the first aspect and its possible designs according to the instructions The described map drawing method.
- Fig. 1 is a kind of schematic diagram of the high-precision positioning map obtained by the method of laser point cloud;
- FIG. 2 is a simplified schematic diagram of a road scene
- FIG. 3 is a schematic flowchart of a map drawing method provided by an embodiment of the present application.
- FIG. 4 is a schematic diagram of a road sign provided by an embodiment of the present application.
- 5A is a schematic diagram of determining a relative height according to an embodiment of the present application.
- FIG. 5B is a schematic diagram of yet another relative height determination provided by an embodiment of the present application.
- FIG. 6 is a schematic diagram of obtaining a positioning picture according to an embodiment of the present application.
- FIG. 7 is a schematic diagram of obtaining another positioning picture provided by an embodiment of the present application.
- FIG. 8 is a schematic diagram of a method for acquiring a local feature provided by an embodiment of the present application.
- FIG. 9 is a schematic diagram of acquiring a characteristic fingerprint according to an embodiment of the present application.
- FIG. 10 is a schematic diagram of a method for determining similarity based on feature fingerprints provided by an embodiment of the present application.
- FIG. 11 is a schematic diagram of a comparison of a group of behavior pictures provided by an embodiment of the present application.
- FIG. 12 is a comparative schematic diagram of another group of positioning pictures provided by an embodiment of the present application.
- FIG. 13 is a comparative schematic diagram of another group of positioning pictures provided by an embodiment of the present application.
- FIG. 14 is a schematic comparison diagram of another group of positioning pictures provided by an embodiment of the present application.
- 15 is a schematic diagram of a fusion picture provided by an embodiment of the present application.
- 16 is a schematic diagram of a map drawing device provided by an embodiment of the present application.
- FIG. 17 is a schematic diagram of another map drawing device provided by an embodiment of the present application.
- FIG. 18 is a schematic diagram of a chip system provided by an embodiment of the present application.
- a high-precision positioning map is a map that can be directly recognized and used by the device and can provide detailed map data.
- a device such as a vehicle
- high-precision positioning maps can be obtained through image and global positioning system (global positioning system, GPS) technology.
- a measuring device such as a robot drives on the road in the area where the map needs to be drawn, and obtains environmental pictures at different locations during the process of driving on the road by taking pictures.
- different positions can be obtained through GPS positioning. According to these environmental pictures and referring to the position of the robot, the corresponding high-precision positioning map can be obtained.
- the high-precision positioning map drawn and obtained according to the image and GPS technology is mainly used in scenarios with low requirements for map accuracy, such as automatic driving level at layer 2 (layer 2, L2) or layer 3 (layer 3, L3).
- map accuracy such as automatic driving level at layer 2 (layer 2, L2) or layer 3 (layer 3, L3).
- the hyperlocation map can be used to support the implementation of advanced driving assistance systems.
- a higher level of automatic driving such as a scene where the level of automatic driving is at layer 4 (layer 4, L4) or layer 5 (layer 5, L5)
- due to the high requirements for map accuracy it cannot be used through the above Imagery and GPS technology to draw high-precision maps.
- a high-precision positioning map with high accuracy can be obtained through the measurement method of laser point cloud.
- a plurality of different measuring devices are respectively driving on a road where a map needs to be drawn, and map information corresponding to different positions during road driving is obtained through components such as a laser measurement module arranged on the robot.
- the map information may include information such as the three-dimensional environment measured by the laser measurement module, echo intensity, and the like.
- the robot's pose such as the robot's position and the angle when the robot obtains the map information
- the corresponding sensor information when it obtains the map information.
- draw the corresponding local map can be obtained by a method of simultaneous localization and mapping (SLAM).
- SLAM simultaneous localization and mapping
- a high-precision positioning map corresponding to the area where the map needs to be drawn can be finally obtained by fusing multiple local maps corresponding to the map information collected by multiple robots.
- the relationship between the two local maps can be determined according to the relationship between the information used to indicate the same location (ie, the location in the overlapping area) in different local maps, and then accurate fusion can be performed.
- a universal coordinate system based on the universal transverse mercator grid system can be used as the unified coordinate system for all robots to collect map information.
- the pose information of different robots can be mapped into the UTM coordinate system through Taylor differential transformation (or T transformation), so that the robot can carry out Fusion of map information.
- the collection of the robot's map information may also be performed according to different coordinate systems.
- FIG. 1 shows a schematic diagram of a high-precision positioning map obtained by a method of laser point cloud.
- the high-precision positioning map can display the distribution of objects (such as trees, houses, roads, etc.) on the horizontal plane in the corresponding area in the form of a top view.
- the pixels corresponding to the location of the area without object distribution can be displayed as black.
- the pixels at the corresponding positions can be displayed in grayscale or white.
- the grayscale can be determined by data such as echo parameters in the map data.
- the road condition information with three-dimensional spatial distribution in the map is compressed into only the horizontal distribution. Therefore, the height information is lost, and the vehicle cannot distinguish the distribution of road conditions at different heights according to the map.
- FIG. 2 As shown in (a) of FIG. 2 , the scene includes a road A, and a road B whose vertical road is higher than the road A.
- the high-precision map obtained by the method of laser point cloud in the current technology is shown in (b) of Fig.
- the distribution relationship means that when a vehicle travels to the projected intersection area of road A and road B, problems such as inability to provide correct positioning information may occur. Especially when the vehicle is in autonomous driving, this kind of localization failure can lead to serious problems such as damage to the vehicle as well as the road.
- the embodiments of the present application provide a map drawing method, which can draw corresponding positioning maps for roads with different heights, so that the device can flexibly locate the maps according to the multiple high-precision positioning maps and combine its own location. Selecting the high-precision positioning map at the corresponding height for positioning can effectively avoid the problem that the device cannot obtain accurate positioning information based on the high-precision positioning map due to the fact that the high-precision positioning map does not include absolute elevation information. It should be understood that, due to the above-mentioned problems such as the inability to distinguish the distribution of roads at different heights in the current map drawing solution, it becomes extremely difficult to construct a large-scale map. However, the map drawing method provided by the embodiment of the present application can effectively solve the above problem, and thus can support a map construction process of a high-precision positioning map in a large area.
- the positioning pictures of the road layers corresponding to different roads can be obtained, and according to the similarity of these positioning pictures, it is determined whether the different positioning pictures are in the layer corresponding to the same road, so as to determine whether the different positioning pictures are in the layer corresponding to the same road.
- the positioning pictures of the same layer are fused, and finally multiple layers of roads distributed at different heights are obtained as a high-precision positioning map.
- map drawing method provided by the embodiments of the present application can be applied to the scene of high-precision positioning map drawing, and in particular, can efficiently provide support for the drawing of high-precision positioning maps used for automatic driving.
- the high-precision positioning map drawn and obtained by this method can also be applied to other scenarios, such as scenarios involving autonomous movement and positioning of intelligent robots.
- the technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.
- the high-precision positioning map may also be referred to as a positioning map for short.
- FIG. 3 is a schematic flowchart of a map drawing method according to an embodiment of the present application. As shown in FIG. 3, the method may include S301-S304.
- the laser point cloud data can be acquired by measuring equipment such as robots and vehicles.
- the following takes the measurement device as a vehicle provided with a laser measurement module (such as a laser sensor) as an example.
- the vehicle can drive on the road in the area that needs to be mapped (such as the area to be mapped), and collect laser point cloud data corresponding to objects in the surrounding environment at different positions during the driving process.
- the laser sensor on the vehicle can emit laser signals and receive corresponding feedback signals.
- the corresponding laser point cloud data can be obtained according to the relevant parameters of the feedback signal.
- the parameters corresponding to the feedback signal corresponding to the laser signal may include the reflected position of the laser signal (such as the position corresponding to the laser point cloud) in the global coordinate system (such as geographic coordinate system, Cartesian coordinate system, etc.) XYZ three-dimensional coordinate information.
- the X and Y coordinates may be used to identify the horizontal coordinates of the position corresponding to the laser point cloud
- the Z coordinate may be used to identify the vertical coordinates of the position corresponding to the laser point cloud.
- the laser point cloud data may also include other information, such as the number of echoes of the laser signal, intensity information, category, scanning angle of the vehicle, and scanning direction and other information.
- the lane marking is more important than other objects in the process of automatic driving of the vehicle.
- the road marking can accurately determine the driving road. Conditions (such as whether to turn, whether there is a crosswalk ahead, number of lanes, etc.). Therefore, in this example, the parameter corresponding to the feedback signal corresponding to the laser signal may also include an indication of whether the position corresponding to the laser signal is a lane marking.
- the lane markings may be lane markings, road signs and other markings. For the convenience of description, the following takes the lane marking as the lane line as an example. For example, as shown in FIG.
- the lane line is the lane mark as shown in FIG. 4 .
- the laser point cloud data collected at the corresponding position of the lane line can be marked as a lane mark.
- the laser point cloud data collected from the corresponding positions of other non-lane lines can be marked with non-lane markings. It should be noted that, in different implementation manners of the present application, the effect of marking a lane marking or marking a non-lane marking in the above example can be achieved by different methods.
- a fixed bit in the laser point cloud data that needs to be marked with lane marks can be set to 1 (or set to 0) to indicate that the corresponding position of the laser point cloud is Lane markings.
- the fixed position 0 (or 1) in the laser point cloud data that needs to be marked with a non-lane mark, or not blank, is used to indicate that the corresponding position of the laser point cloud is not a lane mark.
- the fixed position for identifying whether the corresponding laser point cloud data is a lane mark may also be a convention mark in the laser point cloud data. This embodiment of the present application does not limit the marking method of the lane marking.
- the laser point cloud data corresponding to the position of the lane line is marked with 1
- the laser point cloud data corresponding to the position of the non-lane line is marked with 0 as an example.
- the corresponding features can be flexibly selected according to the needs of the corresponding scenarios to replace the Road signs, in order to be able to more accurately distinguish important features from other features, and then obtain a high-precision positioning map corresponding to the needs of the scene.
- the vehicle can store the laser point cloud data locally in real time, or upload it to the cloud (or server) through the network.
- the network in this example may be the third generation mobile communication technology (3rd-Generation, 3G), the fourth generation mobile communication technology (4th-Generation, 4G), the fifth generation mobile communication technology (5th-Generation, 5G) ) or other network that can be used for data transfer.
- the vehicle can also upload the collected laser point cloud data in a preset cycle. This embodiment of the present application does not limit this.
- the laser point cloud data acquired by one or more vehicles may be preprocessed, so that the laser point cloud data can more accurately characterize the distribution of objects in the area to be drawn.
- the preprocessing may include one or more of processing operations such as denoising, derotation, alignment, and downsampling.
- the road layers at different height layers are drawn respectively to obtain the positioning map of the corresponding height layer. Therefore, in this example, in the process of preprocessing, the height information based on the global coordinate system (such as the Z coordinate in the XYZ coordinates) included in the laser point cloud data can be converted into relative to the collected laser point cloud data.
- the relative height of the road being driven at the time, which satisfies laws and regulations and facilitates subsequent processing.
- FIG. 5A is a schematic diagram of determining a relative height.
- the scene shown in Figure 5A is the collection of laser point cloud data during the process of the vehicle running on the ground.
- the relative height of the laser point cloud data acquired in the acquisition process can be determined by taking the ground as a reference.
- the relative height of point A on the ramp is the distance H1 from the point A to the plane on which the ground is located.
- the relative height of point B on the elevated is the distance H2 from the point B to the plane on which the ground is located.
- the height relative to the closest point of the road edge can be used as its relative height information relative to the ground as shown in FIG. 5A .
- the relative height information of different objects can be determined with reference to the road surface of the ramp.
- FIG. 5B shows a schematic diagram of the determination of the relative height under the XOZ plane.
- the elevated and the ground are two parallel roads. Therefore, the relative height of the laser point cloud data corresponding to the elevated is H3.
- the laser point cloud data corresponding to the overhead includes P3 as shown in Figure 5B, and its relative height is H3.
- the relative height of point P4 on the ramp may be H4, and the relative height of point P5 on the ramp may be H5.
- the laser point cloud data collected by one or more vehicles may be processed to obtain multiple positioning pictures.
- each positioning image can correspond to a part of the image in a road layer on the XOY plane.
- the corresponding positioning picture can be determined according to the laser point cloud data obtained by the same vehicle in a continuous period of time (eg T1). Therefore, it is ensured that the positioning icon piece does not include information in different layers at the same time.
- the length of the T1 depends on the size of the positioning image to be obtained.
- the length of T1 can be determined according to information such as the driving speed of the vehicle, and the laser point cloud data within the T1 period can be selected for fusion to obtain the corresponding positioning picture.
- the sizes of any two positioning pictures among the multiple positioning pictures may be the same or different. The following is an example of the same size of different positioning images.
- the positioning picture when obtaining the positioning picture according to the laser point cloud data, the positioning picture may be limited to one
- the laser point cloud data whose height difference is greater than a certain threshold are not included in the positioning image.
- a positioning image does not include laser point cloud data with a height difference greater than 4 meters.
- each pixel in the positioning picture can correspond to a set of laser point cloud data with the same horizontal coordinates (ie X coordinates and Y coordinates) but with different relative heights .
- the positioning picture can be a grayscale plane map, and the grayscale of each pixel can be determined according to the corresponding laser point cloud data.
- each pixel in the positioning picture may also correspond to a set of laser point cloud data having the same horizontal coordinate range.
- the laser point cloud data corresponding to each pixel may be all laser point cloud data whose X coordinate is in the range of [X1, X2] and the Y coordinate is in the range of [Y1, Y2].
- each pixel in the positioning picture can represent more information corresponding to the laser point cloud data, thereby reducing the number of positioning pictures the goal of.
- the positioning picture is a two-dimensional image in the XOY plane
- the distribution of the corresponding road layer on the XOY plane can be visually displayed on the positioning picture.
- information such as the relative height information of the corresponding position and the road sign can also be identified by using the grayscale information of different pixels.
- the grayscale information of each pixel may correspond to a single-channel multi-bit binary number. Different bits of the binary number can be used to identify information such as the relative height of the object at the corresponding position of the pixel. After converting the binary number to decimal, it can correspond to the grayscale corresponding to the pixel.
- the method for determining the single-channel multi-bit binary number is exemplified below. Take the grayscale information of each pixel corresponding to a single-channel 8-bit binary number as an example.
- the 8-bit binary number is converted into decimal, and the corresponding decimal number can be the grayscale of the corresponding pixel.
- 7 bits (such as the 0th to the 6th position) can be used to identify the height distribution of the laser point cloud in a horizontal coordinate, and the remaining 1 bit (such as the 7th position) can be used to identify the corresponding Whether the object at the location is a lane marker.
- each bit of the 0th bit to the 6th bit of the 8-bit binary number may be filled according to the corresponding relationship shown in Table 1 below.
- Table 1 shows only an example of a corresponding relationship.
- more or better bits may also be used to match multiple laser point clouds in the same horizontal coordinate.
- height distribution for identification The relative height information corresponding to each bit may also be different, which is not limited in this embodiment of the present application.
- the remaining one bit (eg, the seventh bit) in the 8-bit binary number can be used to identify whether the laser point cloud under the horizontal coordinate is a lane mark. For example, when the laser point cloud is a lane marking, set the 7th bit to 1. Correspondingly, when the laser point cloud is not a lane mark, the seventh bit is set to 0.
- the vertical distribution of multiple laser point clouds with the same horizontal coordinates can be determined by 8-bit binary numbers.
- the grayscale of the pixel corresponding to the positioning picture under the horizontal coordinate can be determined according to the 8-bit binary number.
- the 8-bit binary number can be converted into a decimal number as the grayscale of the pixel, and the pixel can be filled with the grayscale, and finally a positioning picture can be obtained.
- FIG. 6 Continuing to take the case of identifying the height distribution by a single-channel 8-bit binary number as an example.
- a tree is included in the horizontal coordinate, and the 0th to 6th bits can be filled by the description in the above method.
- the binary number of 00011111 can be filled and used for Identifies the distribution of heights under this horizontal coordinate.
- the seventh bit of the binary number can be used to identify whether the laser point cloud under the horizontal coordinate is a lane mark. Since the object of the laser point cloud in the figure is a tree, the seventh position can be 0.
- an 8-bit binary number including information on the height distribution and whether it is a lane marking can be obtained.
- the grayscale of the corresponding pixel under the horizontal coordinate can be obtained.
- the grayscales of other pixels in the positioning picture can be determined according to this method, and the corresponding positioning pictures can be obtained by filling them.
- the positioning map and the positioning picture are both grayscale images as an example for description.
- the positioning map to be drawn and the positioning picture are color images (such as RGB images)
- the RGB color spectrum of each pixel can also correspond to a single-channel multi-bit binary number.
- the specific execution method It is similar to the above-mentioned drawing method of the grayscale image, and will not be repeated here.
- the positioning map can be obtained by fusing the multiple positioning pictures.
- the positioning pictures of different road layers may be fused respectively to obtain the positioning map of the corresponding road layer.
- the road layers in the above description may correspond to roads with different heights. That is to say, in this embodiment of the present application, for a road with different heights, a set of positioning pictures corresponding to the road layer of the road may be determined.
- the first road may correspond to the first road layer, and the first road layer may include the positioning pictures formed by the collected laser point cloud data when the vehicle drives on the first road among the above-mentioned multiple positioning pictures. collection.
- the second road may correspond to the second road layer
- the second road layer may include the positioning formed by the collected laser point cloud data when the vehicle is driving on the second road in the above-mentioned multiple positioning pictures. Collection of pictures.
- the absolute height of the first road (that is, the height of the first road in the global coordinate system) may be referred to as the location corresponding to the laser point cloud data collected and acquired during the vehicle traveling on the first road The reference height of the image.
- the absolute height of the second road (that is, the height of the second road in the global coordinate system) can be referred to as the reference of the positioning picture corresponding to the collected laser point cloud data when the vehicle is traveling on the second road. high.
- the reference heights for any two positioning pictures included in the same road layer can be included in a preset range. That is, the reference height difference for any two positioning pictures included in the same road layer does not exceed the first threshold.
- the reference heights for the positioning pictures included in different road layers may not be within the preset range. That is to say, the reference height difference for any two positioning pictures included in different road layers is greater than the second threshold.
- the first threshold may be the same as the second threshold, or may be different from the second threshold. The specific situation can be flexibly selected or set according to actual needs, which is not limited in this embodiment of the present application.
- the road layers corresponding to different positioning pictures in order to determine the road layers corresponding to different positioning pictures, it may be determined whether different positioning pictures are in the same road layer. If the positioning pictures are in different road layers, the corresponding positioning pictures do not need to be fused.
- the method for judging whether multiple positioning pictures covering the same horizontal area are in the same road layer is described below with reference to an example.
- the positioning picture 1 and the positioning picture 2 cover the same horizontal area as an example.
- it can be determined whether two positioning pictures eg, positioning picture 1 and positioning picture 2
- a window sliding method may be used respectively to determine the mean, and/or variance, and/or covariance of local features in each positioning picture.
- the similarity of the two positioning pictures is determined.
- a preset threshold eg, threshold 1
- threshold 1 a preset threshold
- FIG. 8 is a schematic diagram of a method for acquiring a local feature provided by an embodiment of the present application.
- the positioning image includes 6*6 pixels, the size of the window is 3*3, and the sliding step size is 1 as an example.
- the initial position of the window may be located at the upper left corner of the positioning picture, corresponding to pixels covering three rows and three columns in the upper left corner of the positioning picture.
- the gray level average ⁇ 1 of the window, the gray level variance ⁇ 1 and the gray level covariance can be determined.
- the window can be slid to the right by 1 pixel to the position shown in (b) of FIG. 8 . And obtain the gray average ⁇ 2 , the gray variance ⁇ 2 and the gray covariance of the position. After that, you can continue to swipe right to get local features in other locations.
- the window can be slid down by one pixel (for example, moved to the position shown in (c) in Figure 8), and the local features of other positions can be obtained by referring to the above method. This is repeated until the window is moved to the position shown in (d) in FIG. 8 , and the local features of the position are acquired.
- each local feature corresponding to the positioning image is acquired. For example, as shown in Figure 8, a total of 16 sets of local features corresponding to each window position can be obtained.
- Each group of local features includes the average gray level of the corresponding position, the variance of the gray level, and the covariance of the gray level.
- a contrast function of the two positioning images can be constructed for the average gray level, the variance of the gray level, and the covariance of the gray level, respectively.
- the contrast function of the gray-scale average (or referred to as the feature mean) can be evaluated by the following formula (1).
- l(x, y) is the feature mean comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position.
- ⁇ x is the feature mean of the positioning picture 1 corresponding to the position.
- ⁇ y is the feature mean of the positioning picture 2 corresponding to the position.
- C 1 is a constant.
- ⁇ x can be obtained according to the following formula: Among them, H is the height of the window, W is the width of the window, and X(i,j) is the pixel value (such as grayscale) at the corresponding position of the pixel (i,j).
- the contrast function of the variance of the gray level (or called the characteristic variance) can be evaluated by the following formula (2).
- c(x, y) is the feature variance comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position.
- ⁇ x is the feature variance of the positioning picture 1 corresponding to the position.
- ⁇ y is the feature variance of the positioning picture 2 corresponding to the position.
- C2 is a constant.
- ⁇ x can be obtained according to the following formula:
- the contrast function of the covariance of gray levels (or called feature covariance) can be evaluated by the following formula (3).
- s(x, y) is the feature covariance comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position.
- ⁇ xy is the covariance of the corresponding pixel (eg, X(i, j)) in the positioning picture 1, and the corresponding pixel (eg, Y(i, j)) in the positioning picture 2.
- the covariance can be obtained by the following formula:
- the similarity between the two positioning pictures can be determined according to the comparison value.
- the similarity of a certain window position in the two positioning pictures can be determined according to the following formula (4).
- S(x, y) is the similarity between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position.
- the similarity between the whole positioning pictures can be determined according to the statistical method.
- the similarity between the positioning picture 1 and the positioning picture 2 can be determined according to the following formula (5).
- MS(x, y) is the similarity between the positioning picture 1 and the positioning picture 2.
- M is the number of corresponding positions of the local features obtained in each positioning image.
- j is any one of the M positions.
- ⁇ xj is the feature mean when the window is located at the position corresponding to j in the positioning picture 1.
- ⁇ yj is the feature mean value when the window is located at the position corresponding to j in the positioning picture 2.
- ⁇ xj is the feature variance when the window is located at the position corresponding to j in the positioning picture 1.
- ⁇ yj is the feature variance when the window is located at the position corresponding to j in the positioning picture 2.
- ⁇ xjyj is the covariance.
- the positioning picture 1 and the positioning picture 2 are in the same road layer according to the magnitude relationship between MS(x, y) and the first threshold. For example, when MS(x, y) is greater than the first threshold, it means that the similarity between the positioning picture 1 and the positioning picture 2 is high, so the positioning picture 1 and the positioning picture 2 are considered to be the positioning pictures in the same road layer. If MS(x, y) is less than the first threshold, it means that the similarity between positioning picture 1 and positioning picture 2 is low, so it is considered that positioning picture 1 and positioning picture 2 cover the same horizontal area but are located in different road layers Position the picture. As an example, the first threshold may be 0.5.
- the size of the positioning image, the size of the window and the sliding step size are all exemplary descriptions. In some other implementations, the size of the positioning image, the size of the window and the sliding step size, etc.
- the parameters can be flexibly selected according to actual needs, which are not limited in this embodiment of the present application.
- the local features include feature mean, feature variance, and feature covariance at the same time, and the similarity between the positioning picture 1 and the positioning picture 2 is jointly determined with reference to these three parameters. In other implementations, only one or any two of the feature mean, feature variance, and feature covariance may be referred to to determine the similarity between the positioning picture 1 and the positioning picture 2.
- the local texture and global reference method obtains the similarity between the two positioning pictures by performing detailed evaluation and comparison of the local features of the positioning pictures. Therefore, the similarity measurement result can better reflect the difference in the details of the positioning pictures. When the environment complexity corresponding to the positioning pictures is low, the difference between the two positioning pictures can be effectively judged according to the local features. A similarity measure when the environment is simpler.
- the number of pixel features in the positioning image may be large.
- the two positioning pictures can be respectively reduced to a smaller size, and the relative height represented by each pixel in the reduced picture can be calculated and obtained.
- the positioning picture is normalized according to the height mean value, thereby obtaining the respective height features of the two positioning pictures. By comparing the height features of the two positioning pictures, the similarity of the two positioning pictures is determined.
- the two positioning pictures are first processed.
- the size of the positioning picture can be reduced by deleting pixels in every row/column.
- the positioning picture is reduced to a pixel size of 3*3 as shown in (a) of FIG. 9 .
- the pixel in the i-th row and the j-th column in the reduced picture is identified by a ij , where i and j are both positive integers less than or equal to 3.
- the detail information in the positioning image is removed, and only the basic structure and light and shade information are retained, according to the grayscale of each pixel, Determines the relative height of the corresponding pixel.
- the relative height of a 11 is h 11
- the relative height of a 12 is h 12
- the relative height of a ij is h ij .
- the average height ⁇ h of each pixel of the reduced picture can be obtained.
- the corresponding positioning picture can be normalized.
- normalization processing can be performed by the following method: comparing the magnitude relationship between h ij and ⁇ h, if h ij is greater than ⁇ h, the corresponding a ij is marked as 1. Conversely, if h ij is less than ⁇ h, the corresponding a ij is marked as 0. In this way, a 3*3 matrix can be obtained. where each element is either 0 or 1.
- a matrix as shown in (b) in FIG. 9 can be acquired. Arrange the elements in the matrix in order to obtain a 9-bit binary number. Taking the matrix as shown in (b) in FIG. 9 as an example, the corresponding binary number after the sequential arrangement may be 101001111. Thus, the binary number can be called the feature fingerprint of the corresponding positioning picture.
- the feature fingerprints respectively corresponding to the positioning picture 1 and the positioning picture 2 can be obtained.
- the similarity of the two images can be determined.
- the similarity of the two positioning images can be determined by calculating the Hamming distance of the two feature fingerprints.
- the feature fingerprint of the positioning picture 1 and the feature fingerprint of the positioning picture 2 are different by 3 bits as shown in the dotted box in the figure, and the Hamming distance of the feature fingerprints of the two positioning pictures is 3. It can be understood that the larger the Hamming distance, the lower the similarity between the two positioning images. The smaller the Hamming distance, the higher the similarity between the two positioning images.
- the size relationship between the Hamming distance and the second threshold can be compared to determine whether the two positioning pictures are in the same road layer. For example, when the Hamming distance is greater than the second threshold, it is considered that the two positioning images are not in the same road layer. For another example, when the Hamming distance is less than the second threshold, it is considered that the two positioning images belong to the same road layer.
- the second threshold may be 8.
- the similarity determined according to the feature fingerprint may be referred to as the fingerprint similarity (map fingerprint similarity, mfs).
- the map feature fingerprint method can better reflect the global differences of the positioning pictures by filtering out the details in the positioning pictures, retaining the basic information, and measuring the similarity of the positioning pictures from the global perspective. Therefore, it is more suitable for similarity measurement when the environment corresponding to the positioning picture is relatively complex.
- the two similarity measurement methods provided in the above example can be flexibly selected and used according to different scenes, or both methods can be used at the same time. Perform an adaptive similarity measure.
- the parameters used to characterize the complexity of the environment can be introduced to adjust the weight of the similarity determined by the two methods respectively in the result, so that according to the complexity of the environment The degree of similarity is measured adaptively.
- the final similarity can be obtained by the following formula (6).
- MapS f(MS, mfs, ⁇ ) ... Equation (6).
- MapS is the similarity between the positioning picture 1 and the positioning picture 2.
- MS is the similarity determined according to the local texture and global reference method.
- mfs is the similarity determined according to the map feature fingerprinting method.
- ⁇ is the environment complexity.
- the formula (6) can be set flexibly.
- the formula (6) can be transformed into the following formula (6-1).
- ⁇ can be flexibly set to a number less than or equal to 1 and greater than or equal to 0 according to changes in the complexity of the environment.
- the global similarity can be used as the main similarity measure, so ⁇ can be set in the range of (0.5, 1], which can improve the weight of MS.
- ⁇ can be set in the range of [0, 0.5), which can increase the weight of mfs. In this way, the purpose of adaptively adjusting the similarity measure of the positioning pictures according to the different complexity of the environment can be achieved.
- the MapS can be compared with the third threshold to determine whether the positioning picture 1 and the positioning picture 2 are in the same road layer. For example, if MapS is greater than the third threshold, it is considered that the similarity between the positioning picture 1 and the positioning picture 2 is high and belong to the same road layer. If MapS is less than the third threshold, it is considered that the similarity between the positioning picture 1 and the positioning picture 2 is low, and they belong to different road layers.
- the positioning pictures covering the same horizontal area can be used as the object of similarity measurement, so as to determine whether different positioning pictures are collected by vehicles on the same road at different times.
- any two positioning pictures in the multiple positioning pictures can also be used as the objects of similarity measurement, so that through this method, it can be accurately determined whether any two positioning pictures in the multiple positioning pictures are in the same one road layer.
- the environment features are more complex, and the positioning picture 1 and positioning picture 2 are based on the laser point cloud data obtained in the same road layer and collected at different times as an example.
- the positioning picture 1 and the positioning picture 2 are shown in the figure.
- the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 3
- the similarity MS obtained based on the local texture and the global reference method is 0.6192.
- the weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.785823.
- the environmental characteristics are relatively simple, and the positioning picture 1 and the positioning picture 2 are obtained according to the laser point cloud data collected at different times in the same road layer as an example.
- the positioning picture 1 and the positioning picture 2 are shown in the figure.
- the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 1, and the similarity MS obtained based on the local texture and the global reference method is 0.885771.
- the weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.927816.
- the environment features are more complex, and the positioning picture 1 and positioning picture 2 are obtained from laser point cloud data collected from different road layers as an example.
- the positioning picture 1 and the positioning picture 2 are shown in the figure.
- the Hamming distance calculated to obtain the feature fingerprints corresponding to the two positioning pictures is 13
- the similarity MS obtained based on the local texture and the global reference method is 0.051161.
- the weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.085639.
- the environment features are relatively simple, and the positioning picture 1 and the positioning picture 2 are obtained from the laser point cloud data collected from different road layers as an example.
- the positioning picture 1 and the positioning picture 2 are shown in the figure.
- the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 0, and the similarity MS obtained based on the local texture and the global reference method is 0.133746.
- the weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.165378.
- the weight of the similarity obtained by the two methods can be adjusted reasonably, such as increasing the weight of the similarity obtained based on the local texture and the global reference method, so as to obtain the similarity closer to the actual Measure the results.
- the similarity obtained based on the local texture and global reference method is large.
- the premise of the two experiments limits the positioning picture 1 and the positioning picture 2 to be the positioning pictures in the same road layer. Therefore, when the environmental features are relatively simple, the similarity obtained based on the map feature fingerprinting method can more accurately reflect the similarity of the two positioning pictures. That is to say, when the environmental features are relatively simple, if a single method is used to measure the similarity, the similarity obtained by the map feature fingerprinting method can be prioritized to judge whether the two positioning pictures are in the same road layer.
- the weight of the similarity obtained by the two methods can be adjusted reasonably, such as increasing the weight of the similarity obtained by the map feature fingerprinting method, so as to obtain the similarity measurement result that is closer to the actual.
- the two positioning pictures may be fused to obtain a positioning picture with more accurate detailed information.
- the two pictures in Experiment 1 can be fused.
- the fusion result is shown in (a) of FIG. 15 . Comparing the positioning picture shown in (a) in Figure 15 with the two positioning pictures in Experiment 1, it can be seen that the fused positioning picture has more accurate detailed information. Therefore, the fusion process helps to compensate for the single-shot The collected map is missing.
- the result of merging the two pictures in Experiment 2 is shown in (b) in Figure 15, which also has similar characteristics compared with the pictures before fusion.
- the positioning pictures in the same road layer can be fused to obtain the corresponding road map The location map of the layer.
- different map labels (such as map id) can be added to different road layers, so that during the automatic driving process of the vehicle, it can be based on the current position information (such as the current global location determined by the GPS positioning system). height information in the coordinate system), select the positioning map corresponding to the map id to obtain accurate positioning information.
- the fusion process may include various specific implementations.
- two pixels at each corresponding position in the two positioning pictures may be fused to Get a fused pixel.
- the pixel values of the two pixels can be averaged as the pixel value of the pixel after fusion. It is also possible to take the maximum value of the pixel values of two pixels as the pixel value of the pixel after fusion. Of course, the minimum value of the pixel values of the two pixels can also be taken as the pixel value of the pixel after fusion.
- the embodiments of the present application do not limit the pixel fusion processing mechanism.
- the set of the first positioning picture can be respectively Fusion processing is performed on the first positioning picture in the first road layer to obtain a positioning map corresponding to the first road layer.
- the fusion processing of positioning maps with different horizontal coverage areas may be performed according to the horizontal coordinates corresponding to the pixels at the edge in the positioning picture.
- the horizontal plane coordinates of the pixel in the lower left corner of the positioning picture A in the global coordinate system are (Xa, Ya)
- the positioning picture B whose pixel coordinates in the lower right corner are (Xa, Ya)
- the fusion process may be to place the positioning picture A on the adjacent right side of the positioning picture B to obtain a new positioning picture that includes the positioning picture A and the positioning picture B at the same time.
- the set of positioning pictures is not fused, but the set of positioning pictures is stored in the server, or sent to the server.
- equipment eg vehicles.
- the vehicle needs to use the map of the corresponding road layer, it combines its own position, fuses the positioning pictures in the corresponding road layer that are adjacent to the vehicle's location, and then goes to the local positioning map corresponding to the road layer. In order to provide accurate positioning information for the current driving of the vehicle.
- each functional module may be divided into each function, or two or more functions may be integrated into one processing module.
- the above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. It should be noted that, the division of modules in the embodiments of the present application is schematic, and is only a logical function division, and there may be other division manners in actual implementation.
- FIG. 16 is a schematic diagram of the composition of a map drawing apparatus 1600 according to an embodiment of the present application.
- the map drawing apparatus 1600 may be a terminal, or a chip inside the terminal, and may implement the optional embodiments as shown in FIG. 3 and the above.
- the map drawing apparatus may include: an acquisition unit 1601 and a fusion unit 1602 .
- the obtaining unit 1601 may be configured to perform any one of the steps S301 to S303 in the method shown in FIG. 3 and any one of the optional embodiments.
- the fusion unit 1602 may be configured to perform any of the steps in S304 shown in FIG. 3 and any of the optional embodiments.
- the acquiring unit 1601 is configured to acquire laser point cloud data of a first area to be drawn, where the first area to be drawn includes a first road and a second road on different planes.
- the obtaining unit 1601 is further configured to obtain a set of first positioning pictures and a set of second positioning pictures according to the laser point cloud data.
- the difference between the reference heights of any two first positioning pictures does not exceed the first threshold.
- the difference between the reference heights of any two second positioning pictures does not exceed the first threshold.
- the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold.
- the reference height of the positioning picture is the height of the road corresponding to the device when the laser point cloud data corresponding to the positioning picture is collected. Both the first threshold and the second threshold are positive numbers.
- the fusion unit 1602 is configured to fuse the positioning pictures in the first positioning picture set to obtain a map corresponding to the first road, and fuse the positioning pictures in the second positioning picture set to obtain a map corresponding to the second road.
- the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected.
- the obtaining unit 1601 is further configured to obtain, according to the three-dimensional coordinate information of the first position, the relative height information of the first position relative to the road on which the laser point cloud data is collected.
- the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
- the acquiring unit 1601 is configured to acquire a plurality of positioning pictures according to the laser point cloud data, and the collection time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture The pixel value of the pixel included in the picture is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position.
- the fusion unit 1602 is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set and a second positioning picture set.
- the apparatus further includes: a determining unit 1603, where the determining unit 1603 is configured to determine the similarity between the first positioning picture and the second positioning picture.
- the determining unit 1603 is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, where the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture.
- the first positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures.
- the fusion unit 1602 is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
- the apparatus further includes: a determining unit 1603, the determining unit 1603 is configured to determine the first positioning picture and the second positioning picture according to the local features of the first positioning picture and the second positioning picture. similarity.
- the local features include one or more of the following: the grayscale mean of the pixels in the location image, the grayscale variance of the pixels, and the grayscale covariance of the pixels.
- the determining unit 1603 is further configured to determine that the first positioning picture and the second positioning picture are in the same road layer when the first similarity is greater than the first threshold.
- the determining unit 1603 is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
- the determining unit 1603 is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture. When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
- the determining unit 1603 is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
- the obtaining unit 1601 is specifically configured to perform the following operations on the first positioning picture and the second positioning picture respectively, so as to obtain the feature fingerprint corresponding to the first positioning picture and the feature fingerprint corresponding to the second positioning picture : Delete the pixels of the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the mean value of the relative heights of each pixel in the reduced positioning picture, According to each pixel value of the normalized reduced positioning picture, the feature fingerprint corresponding to the positioning picture is determined.
- the second similarity between the first positioning picture and the second positioning picture is determined, and the second similarity is the feature fingerprint of the first positioning picture and the second positioning picture.
- the Hamming distance of the characteristic fingerprint is the Hamming distance of the characteristic fingerprint.
- the map drawing device in the embodiment of the present application may be implemented by software, for example, a computer program or instruction having the above-mentioned functions, and the corresponding computer program or instruction may be stored in the internal memory of the terminal, and read by the processor.
- the above-mentioned functions are realized by fetching the corresponding computer programs or instructions inside the memory.
- the map drawing apparatus in this embodiment of the present application may also be implemented by hardware.
- the acquiring unit 1601 and/or the fusion unit 1602 and/or the determining unit 1603 may implement their corresponding functions through a processor (eg, an NPU, a GPU, or a processor in a system chip).
- the map drawing apparatus in this embodiment of the present application may also be implemented by a combination of a processor and a software module.
- the obtaining unit 1601 may be an interface circuit of the processor.
- the interface circuit may transmit the acquired multiple laser point cloud data to the processor.
- the processor can be used to preprocess the laser point cloud data from the interface circuit (such as performing S302 shown in FIG. 3 and any possible operation in this step), and the processor can also be used to The laser point cloud data obtains multiple positioning pictures (such as performing S303 as shown in FIG. 3 and any possible operation in this step), and the processor can also be used to obtain different road layers corresponding to the multiple positioning pictures. (for example, performing S304 as shown in FIG. 3 and any possible operation in this step).
- the processor may also be used to perform other operations in the foregoing embodiments, so as to implement any map drawing method provided by the embodiments of the present application.
- FIG. 17 shows a schematic diagram of the composition of a map drawing apparatus 1700 .
- the map drawing apparatus 1700 may include: a processor 1701 and a memory 1702 .
- the memory 1702 is used to store computer-implemented instructions.
- the communication device 1700 can be caused to execute the map drawing method shown in any one of the foregoing embodiments.
- FIG. 18 shows a schematic composition diagram of a chip system 1800 .
- the chip system 1800 may include: a processor 1801 and a communication interface 1802, which are used to support related devices to implement the functions involved in the above embodiments.
- the chip system further includes a memory for storing necessary program instructions and data of the terminal.
- the chip system may be composed of chips, or may include chips and other discrete devices.
- the communication interface 1802 may also be referred to as an interface circuit.
- the functions or actions or operations or steps in the above embodiments may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
- a software program When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product.
- the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, all or part of the processes or functions described in the embodiments of the present application are generated.
- the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
- the computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line, DSL) or wireless (eg, infrared, wireless, microwave, etc.).
- the computer-readable storage medium can be any available medium that can be accessed by a computer, or data storage devices including one or more servers, data centers, etc. that can be integrated with the medium.
- the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)), and the like.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Electromagnetism (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
Disclosed are a map drawing method and apparatus, which relate to the field of image processing, and solve the problem of an existing location map being unable to provide accurate location information. The specific solution is as follows: acquiring laser point cloud data in a first area to be subjected to drawing, wherein said area comprises a first road and a second road located on different planes (S301); according to the laser point cloud data, acquiring a first set of location pictures and a second set of location pictures (S303); and fusing location pictures in the first set of location pictures to obtain a map corresponding to the first road, and fusing location pictures in the second set of location pictures to obtain a map corresponding to the second road (S304).
Description
本申请要求于2020年9月1日提交国家知识产权局、申请号为202010902739.7、申请名称为“一种地图绘制方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。This application claims the priority of the Chinese patent application with the application number 202010902739.7 and the application title "A method and device for mapping maps", which was submitted to the State Intellectual Property Office on September 1, 2020, the entire contents of which are incorporated herein by reference. middle.
本申请实施例涉及图像处理领域,尤其涉及一种地图绘制方法及装置。The embodiments of the present application relate to the field of image processing, and in particular, to a map drawing method and device.
随着自动驾驶技术的发展,为进行自动驾驶的设备提供定位信息的高精度定位地图,也被广泛地关注。区别于常规地图,高精度定位地图能够提供更加详细的地图数据,并且能够直接被设备(如进行自动驾驶的车辆)识别并使用,由此为车辆的自动驾驶提供精确的定位信息。可以理解的是,高精度定位地图的精度越高,能够为车辆提供的定位信息就越准确,更加有利于车辆的自动驾驶。With the development of autonomous driving technology, high-precision positioning maps that provide positioning information for autonomous driving equipment have also received extensive attention. Different from conventional maps, high-precision positioning maps can provide more detailed map data, and can be directly identified and used by devices (such as vehicles for autonomous driving), thereby providing accurate positioning information for autonomous driving of vehicles. It can be understood that the higher the accuracy of the high-precision positioning map, the more accurate the positioning information can be provided for the vehicle, which is more conducive to the automatic driving of the vehicle.
然而,由于法规的约束,目前的高精度地图无法提供准确的高度信息,由此就会使得车辆无法根据当前的高精度地图区分在不同高度上有不同道路分布(如城市中的立交等)的情况下的路况,这也就使得车辆的自动驾驶很容易出现问题,例如车辆无法在复杂的环境下实现精确的定位。However, due to the constraints of laws and regulations, the current high-precision map cannot provide accurate height information, which makes it impossible for vehicles to distinguish between roads with different road distributions at different heights (such as interchanges in cities, etc.) according to the current high-precision map. This also makes the automatic driving of the vehicle prone to problems, such as the vehicle cannot achieve precise positioning in a complex environment.
发明内容SUMMARY OF THE INVENTION
本申请实施例提供一种地图绘制方法及装置,解决了现有的定位地图无法提供准确的定位信息的问题。The embodiments of the present application provide a map drawing method and device, which solve the problem that the existing positioning map cannot provide accurate positioning information.
为了达到上述目的,本申请实施例采用如下技术方案:In order to achieve the above purpose, the embodiment of the present application adopts the following technical solutions:
第一方面,提供一种地图绘制方法,该方法包括:获取第一待绘制区域的激光点云数据,第一待绘制区域包括处于不同平面的第一道路和第二道路;根据激光点云数据,获取第一定位图片的集合和第二定位图片的集合;其中,各个集合(如第一定位图片的集合,或者第二定位图片的集合)中任意两个第一定位图片的参考高度相差不超过第一阈值;任意两个第二定位图片的参考高度相差不超过第一阈值;第一定位图片和第二定位图片的参考高度差大于第二阈值;定位图片的参考高度是采集定位图片对应的激光点云数据时,对应设备所在路面的高度;第一阈值和第二阈值均为正数;将第一定位图片集合中的定位图片进行融合,获得第一道路对应的地图,将第二定位图片集合中的定位图片进行融合,获得第二道路对应的地图。A first aspect provides a map drawing method, the method comprising: acquiring laser point cloud data of a first area to be drawn, where the first area to be drawn includes a first road and a second road on different planes; according to the laser point cloud data , obtain the set of the first positioning picture and the set of the second positioning picture; wherein, the reference heights of any two first positioning pictures in each set (such as the set of the first positioning picture, or the set of the second positioning picture) are the same Exceed the first threshold; the reference height difference of any two second positioning pictures does not exceed the first threshold; the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is the corresponding height of the collected positioning picture. When the laser point cloud data is obtained, it corresponds to the height of the road where the device is located; the first threshold and the second threshold are both positive numbers; the positioning pictures in the first positioning picture set are fused to obtain the map corresponding to the first road, and the second The positioning pictures in the positioning picture set are fused to obtain a map corresponding to the second road.
基于该方案,提供了一种能够解决不同高度下的定位信息无法被准确判断的问题。示例性的,可以根据处于不同高度的道路,绘制对应的定位地图,以便根据车辆当前所处道路,参考对应道路图层下的定位地图获取定位信息。在本方案中,可以分别获取处于不同道路图层的第一道路和第二道路的定位图片的集合,每个道路图层对应的定位图片的集合即对应该道路图层的定位地图。其中,不同道路图层可以根据该获取该道路图层时,对应设备(如采集激光点云数据的车辆)所在路面在全局坐标系下的绝对高度(即对应定位图片的参考高度)进行区分。例如,采集不同激光点云数据时,车辆所处道路的绝对高度差小于第一阈值,则认为该不同激光点云数据对应的定位图 片处于相同的道路图层中,即该不同的激光点云数据对应的定位图片处于相同的定位图片集合中。反之,采集不同激光点云数据时,车辆所处道路的绝对高度差大于第二阈值,则认为该不同激光点云数据对应的定位图片处于不同的道路图层中,即该不同的激光点云数据对应的定位图片处于不同的定位图片集合中。在不同的实现方式中,该第一阈值可以与第二阈值相同,第二阈值也可以大于第一阈值。该阈值的设置可以根据实际情况灵活选取。由此,即可在不同的道路场景中,特别是具有复杂高度分布的道路场景中,为车辆等需要定位的设备提供准确的定位信息。Based on the solution, a solution is provided that can solve the problem that the positioning information at different heights cannot be accurately judged. Exemplarily, a corresponding positioning map may be drawn according to roads at different heights, so as to obtain positioning information by referring to the positioning map under the corresponding road layer according to the road where the vehicle is currently located. In this solution, sets of positioning pictures of the first road and the second road in different road layers can be obtained respectively, and the set of positioning pictures corresponding to each road layer is the positioning map corresponding to the road layer. Among them, different road layers can be distinguished according to the absolute height of the road surface under the global coordinate system (that is, the reference height of the corresponding positioning picture) where the corresponding device (such as the vehicle that collects laser point cloud data) is located when the road layer is obtained. For example, when collecting different laser point cloud data, if the absolute height difference of the road where the vehicle is located is less than the first threshold, it is considered that the positioning pictures corresponding to the different laser point cloud data are in the same road layer, that is, the different laser point clouds The positioning pictures corresponding to the data are in the same positioning picture set. On the contrary, when collecting different laser point cloud data, if the absolute height difference of the road where the vehicle is located is greater than the second threshold, it is considered that the positioning pictures corresponding to the different laser point cloud data are in different road layers, that is, the different laser point clouds. The positioning pictures corresponding to the data are in different positioning picture sets. In different implementations, the first threshold may be the same as the second threshold, and the second threshold may also be greater than the first threshold. The setting of the threshold can be flexibly selected according to the actual situation. In this way, in different road scenarios, especially in road scenarios with complex height distribution, accurate positioning information can be provided for devices that need to be positioned, such as vehicles.
在一种可能的设计中,该激光点云数据包括:第一位置在全局坐标系下的三维坐标信息,其中,第一位置为采集激光点云数据的位置;在根据激光点云数据,获取第一定位图片集合和第二定位图片集合之前,方法还包括:根据第一位置的三维坐标信息,获取第一位置相对于采集激光点云数据时所行驶道路的相对高度信息。基于该方案,通过相对高度信息标识场景中不同位置(如第一位置)的高度信息。由此即可在合规的前提下,实现对于场景中物体高度的标注。示例性的,根据激光点云数据中的绝对高度信息,结合采集该激光点云数据时,车辆所形式路面的绝对高度,即可确定对应的相对高度信息。In a possible design, the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; Before the first positioning picture set and the second positioning picture set, the method further includes: according to the three-dimensional coordinate information of the first position, obtaining relative height information of the first position relative to the road on which the laser point cloud data is collected. Based on this solution, the height information of different positions (eg, the first position) in the scene is identified by the relative height information. In this way, under the premise of compliance, the height of objects in the scene can be marked. Exemplarily, according to the absolute height information in the laser point cloud data, combined with the absolute height of the road surface in the form of the vehicle when the laser point cloud data is collected, the corresponding relative height information can be determined.
在一种可能的设计中,该激光点云数据还包括:用于指示第一位置是否为车道的第一标识。基于该方案,通过该第一标识,确定第一位置是否为车道标识,以便在后续确定的定位地图,可以据此提供更加准确的路面信息,例如,当前路面的车道标识的位置。In a possible design, the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane. Based on this solution, it is determined whether the first position is a lane mark through the first mark, so that more accurate road surface information can be provided accordingly in the subsequently determined positioning map, for example, the position of the lane mark on the current road surface.
在一种可能的设计中,该在根据激光点云数据,获取第一定位图片集合和第二定位图片集合,包括:根据激光点云数据,获取多个定位图片,每个定位图片对应的激光点云数据的采集时间在预设范围之内,每个定位图片中包括的像素的像素值由像素对应位置的相对高度信息和位置对应的激光点云数据的第一标识确定;将多个定位图片中,处于相同道路图层的定位图片进行融合,获取第一定位图片集合和第二定位图片集合。基于该方案,提供了一种根据激光点云数据获取定位图片的方法。示例性的,可以通过将具有相同水平坐标(如全局坐标系下的XY坐标)的激光点云数据压缩成具有不同像素值的像素点,以据此获取能够表征物体在水平方向的二维分布的定位地图。需要说明的是,根据本示例中的方案,由于每个像素点的像素值(如灰度值)是通过该水平位置不同高度下物体的分布情况确定,因此,根据本示例获取的定位图片,也可以经过一定的处理,还原出在对应区域中的物体的三维分布情况。由此可以在后续定位图片的使用过程中,通过管控上述还原三维分布的处理的方法,使得在不同地区,在合规的前提下能够提供更详细的定位信息。In a possible design, obtaining the first positioning picture set and the second positioning picture set according to the laser point cloud data includes: obtaining a plurality of positioning pictures according to the laser point cloud data, and the laser beam corresponding to each positioning picture The collection time of the point cloud data is within the preset range, and the pixel value of the pixel included in each positioning picture is determined by the relative height information of the corresponding position of the pixel and the first identification of the laser point cloud data corresponding to the position; In the picture, the positioning pictures in the same road layer are fused to obtain the first positioning picture set and the second positioning picture set. Based on this solution, a method for obtaining positioning pictures according to laser point cloud data is provided. Exemplarily, the laser point cloud data with the same horizontal coordinates (such as XY coordinates in the global coordinate system) can be compressed into pixel points with different pixel values, so as to obtain the two-dimensional distribution that can characterize the object in the horizontal direction. location map. It should be noted that, according to the solution in this example, since the pixel value (such as gray value) of each pixel is determined by the distribution of objects at different heights at the horizontal position, the positioning picture obtained according to this example, It is also possible to restore the three-dimensional distribution of objects in the corresponding area after certain processing. In this way, in the subsequent use of the positioning picture, by controlling the above processing method of restoring the three-dimensional distribution, more detailed positioning information can be provided in different regions under the premise of compliance.
在一种可能的设计中,该获取第一定位图片集合和第二定位图片集合,包括:确定第一定位图片和第二定位图片的相似度,并根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,相似度用于指示第一定位图片和第二定位图片的相似程度;其中,第一定位图片和第二定位图片为多个定位图片中,任意两个具有相同水平覆盖区域的定位图片;将多个定位图片中,处于相同道路图层的定位图片进行融合,以获取处于第一道路对应的第一定位图片集合,以及第二道路对应的第二定位图片集合。基于该方案,提供了一种可能的确定处于相同道路图层的定位图片的方案。 示例性的,可以在根据激光点云数据获取多个定位图片后,确定具有相同水平覆盖区域的第一图片和第二图片的相似度,并根据相似度确定该两个定位图片是否是处于相同道路图层中的定位图片。示例性的,当相似度高于对应的预设阈值时,则确定这两个定位图片是处于相同道路图层中的定位图片。对应的,当相似度低于对应的预设预支时,则确定这两个定位图片是处于不同道路图层中的定位图片。In a possible design, acquiring the first positioning picture set and the second positioning picture set includes: determining the similarity between the first positioning picture and the second positioning picture, and determining the first positioning picture and the second positioning picture according to the similarity. Whether the two positioning pictures are in the same road layer, the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture; wherein, the first positioning picture and the second positioning picture are multiple positioning pictures, any two have The positioning pictures of the same horizontal coverage area; the positioning pictures in the same road layer among the multiple positioning pictures are fused to obtain the first positioning picture set corresponding to the first road and the second positioning picture corresponding to the second road gather. Based on this solution, a possible solution for determining positioning pictures in the same road layer is provided. Exemplarily, after obtaining multiple positioning pictures according to the laser point cloud data, the similarity between the first picture and the second picture with the same horizontal coverage area can be determined, and whether the two positioning pictures are in the same position can be determined according to the similarity. The location image in the roads layer. Exemplarily, when the similarity is higher than the corresponding preset threshold, it is determined that the two positioning pictures are positioning pictures in the same road layer. Correspondingly, when the similarity is lower than the corresponding preset advance, it is determined that the two positioning pictures are positioning pictures in different road layers.
在一种可能的设计中,该确定第一定位图片和第二定位图片的相似度,包括:根据第一定位图片和第二定位图片的局部特征,确定第一定位图片和第二定位图片的第一相似度;局部特征包括以下中的一项或多项:定位图片中像素的灰度平均值,像素的灰度方差,像素的灰度协方差;根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,包括:当第一相似度大于第一阈值时,第一定位图片和第二定位图片处于相同道路图层。基于该方案,提供了一种可能的确定相似度的方案,即基于局部特征对比确定两个定位图片的相似度。该方案中,能够对局部特征的差异进行准确的评估,进而获取对应的相似度,因此,能够有效地对定位图片中包括的场景较为简单的定位图片进行较为准确的相似度度量。In a possible design, the determining the similarity between the first positioning picture and the second positioning picture includes: determining the similarity between the first positioning picture and the second positioning picture according to local features of the first positioning picture and the second positioning picture The first similarity; the local features include one or more of the following: the average value of the grayscale of the pixels in the positioning picture, the grayscale variance of the pixels, and the grayscale covariance of the pixels; according to the similarity, determine the first positioning picture and Whether the second positioning picture is in the same road layer includes: when the first similarity is greater than the first threshold, the first positioning picture and the second positioning picture are in the same road layer. Based on this solution, a possible solution for determining the similarity is provided, that is, determining the similarity between two positioning pictures based on the comparison of local features. In this solution, the difference of local features can be accurately evaluated, and then the corresponding similarity can be obtained. Therefore, a relatively accurate similarity measurement can be effectively performed on the positioning pictures with relatively simple scenes included in the positioning pictures.
在一种可能的设计中,该方法还包括:当第一相似度小于第一阈值时,第一定位图片和第二定位图片处于不同道路图层。基于该方案,提供了一种确定两个定位图片不在相同的道路图层的方案。即根据第一相似度的与第一阈值的大小关系,确定第一定位图片和第二定位图片不在相同的道路图层。In a possible design, the method further includes: when the first similarity is less than the first threshold, the first positioning picture and the second positioning picture are in different road layers. Based on this solution, a solution for determining that the two positioning images are not in the same road layer is provided. That is, according to the magnitude relationship between the first similarity and the first threshold, it is determined that the first positioning picture and the second positioning picture are not in the same road layer.
在一种可能的设计中,该确定第一定位图片和第二定位图片的相似度,包括:根据第一定位图片和第二定位图片中对应像素的相对高度信息,确定第一定位图片和第二定位图片的第二相似度;根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,包括:当第二相似度小于第二阈值时,第一定位图片和第二定位图片处于相同道路图层。基于该方案,提供了有一种可能的方案,使得能够准确地获取两个定位图片的相似度。在该示例中,可以根据不同像素点对应的相对高度信息确定相似度。应当理解的是,不同像素的像素值可以是根据对应水平位置的物体的相对高度信息确定,因此,在一些实现方式中,可以根据像素的像素值(如灰度值)确定第一定位图片和第二定位图片的相似度。In a possible design, determining the similarity between the first positioning picture and the second positioning picture includes: determining the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture. The second similarity of the two positioning pictures; according to the similarity, determining whether the first positioning picture and the second positioning picture are in the same road layer, including: when the second similarity is less than the second threshold, the first positioning picture and the second positioning picture are in the same road layer. The positioning image is in the same road layer. Based on this solution, a possible solution is provided so that the similarity between two positioning pictures can be accurately obtained. In this example, the similarity can be determined according to the relative height information corresponding to different pixel points. It should be understood that the pixel values of different pixels may be determined according to the relative height information of the object corresponding to the horizontal position. Therefore, in some implementations, the first positioning picture and The similarity of the second positioning picture.
在一种可能的设计中,该方法还包括:当第二相似度大于第二阈值时,第一定位图片和第二定位图片处于不同道路图层。基于该方案,提供了又一种确定两个定位图片不在相同的道路图层的方案。In a possible design, the method further includes: when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers. Based on this solution, another solution for determining that the two positioning pictures are not in the same road layer is provided.
在一种可能的设计中,该根据第一定位图片和第二定位图片中对应像素的相对高度信息,确定第一定位图片和第二定位图片的第二相似度,包括:针对第一定位图片和第二定位图片分别执行以下操作,以获取第一定位图片对应的特征指纹,和第二定位图片对应的特征指纹:删除定位图片中预设行和/或预设列的像素,以获取缩小的定位图片,根据缩小的定位图片中,各个像素的相对高度的均值,对缩小的定位图片进行归一化处理,根据归一化处理后的缩小的定位图片的各个像素值,确定定位图片对应的特征指纹;根据第一定位图片的特征指纹和第二定位图片的特征指纹,确定第一定位图片和第二定位图片的第二相似度,第二相似度为第一定位图片的特征指纹和第二定位图片的特征指纹的汉明距离。基于该方案,明确了一种根据相对高度信息确定 相似度的可能的实现方式。可以看到,该示例中的特征指纹是能够较为准确地体现定位图片的全局信息,因此,能够通过全局状态下相似度的情况,确定两个定位地图是否处于相同的道路图层。在一些实现场景中,该方案能够更好地对具有较为复杂的定位图片进行准确的相似度度量。In a possible design, determining the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture, including: for the first positioning picture Perform the following operations respectively with the second positioning picture to obtain the feature fingerprint corresponding to the first positioning picture, and the feature fingerprint corresponding to the second positioning picture: delete the pixels of the preset row and/or preset column in the positioning picture to obtain the reduced According to the average value of the relative height of each pixel in the reduced positioning picture, normalize the reduced positioning picture, and determine the corresponding positioning picture according to the pixel values of the reduced positioning picture after normalization. according to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, determine the second similarity of the first positioning picture and the second positioning picture, and the second similarity is the feature fingerprint of the first positioning picture and the second similarity. The Hamming distance of the feature fingerprint of the second positioning image. Based on this scheme, a possible implementation of determining similarity based on relative height information is clarified. It can be seen that the feature fingerprint in this example is the global information that can more accurately reflect the positioning picture. Therefore, it can be determined whether the two positioning maps are in the same road layer through the similarity in the global state. In some implementation scenarios, this solution can better measure the similarity of complex positioning pictures.
第二方面,提供一种地图绘制装置,该装置包括:获取单元,融合单元。该获取单元,用于获取第一待绘制区域的激光点云数据,第一待绘制区域包括处于不同平面的第一道路和第二道路;该获取单元,还用于根据激光点云数据,获取第一定位图片的集合和第二定位图片的集合;其中,任意两个第一定位图片的参考高度相差不超过第一阈值;任意两个第二定位图片的参考高度相差不超过第一阈值;第一定位图片和第二定位图片的参考高度差大于第二阈值;定位图片的参考高度是采集定位图片对应的激光点云数据时,对应设备所在路面的高度;第一阈值和第二阈值均为正数;融合单元,用于将第一定位图片集合中的定位图片进行融合,获得第一道路对应的地图,将第二定位图片集合中的定位图片进行融合,获得第二道路对应的地图。In a second aspect, a map drawing device is provided, the device comprising: an acquisition unit and a fusion unit. The acquisition unit is used to acquire the laser point cloud data of the first area to be drawn, and the first area to be drawn includes the first road and the second road on different planes; the acquisition unit is also used to acquire the laser point cloud data according to the laser point cloud data. The set of the first positioning picture and the set of the second positioning picture; wherein, the reference height difference of any two first positioning pictures does not exceed the first threshold; the reference height difference of any two second positioning pictures does not exceed the first threshold; The reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is the height of the road where the corresponding device is located when the laser point cloud data corresponding to the positioning picture is collected; the first threshold and the second threshold are both is a positive number; the fusion unit is used to fuse the positioning pictures in the first positioning picture set to obtain a map corresponding to the first road, and fuse the positioning pictures in the second positioning picture set to obtain a map corresponding to the second road .
在一种可能的设计中,该激光点云数据包括:第一位置在全局坐标系下的三维坐标信息,其中,第一位置为采集激光点云数据的位置;该获取单元,还用于根据第一位置的三维坐标信息,获取第一位置相对于采集激光点云数据时所行驶道路的相对高度信息。In a possible design, the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; The three-dimensional coordinate information of the first position is obtained, and the relative height information of the first position relative to the road on which the laser point cloud data is collected is obtained.
在一种可能的设计中,该激光点云数据还包括:用于指示第一位置是否为车道的第一标识。In a possible design, the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
在一种可能的设计中,该获取单元,用于根据激光点云数据,获取多个定位图片,每个定位图片对应的激光点云数据的采集时间在预设范围之内,每个定位图片中包括的像素的像素值由像素对应位置的相对高度信息和位置对应的激光点云数据的第一标识确定;该融合单元,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,获取第一定位图片集合和第二定位图片集合。In a possible design, the acquisition unit is configured to acquire multiple positioning pictures according to the laser point cloud data, the acquisition time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture The pixel value of the pixel included in the pixel value is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position; the fusion unit is used to combine the positioning pictures in the same road layer among the multiple positioning pictures Perform fusion to obtain the first positioning picture set and the second positioning picture set.
在一种可能的设计中,该装置还包括:确定单元,该确定单元用于确定第一定位图片和第二定位图片的相似度。该确定单元,还用于根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,相似度用于指示第一定位图片和第二定位图片的相似程度;其中,第一定位图片和第二定位图片为多个定位图片中,任意两个具有相同水平覆盖区域的定位图片。融合单元,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,以获取处于第一道路对应的第一定位图片集合,以及第二道路对应的第二定位图片集合。In a possible design, the apparatus further includes: a determining unit, where the determining unit is configured to determine the similarity between the first positioning picture and the second positioning picture. The determining unit is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, and the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture; wherein the first positioning picture and the second positioning picture are similar. The positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures. The fusion unit is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
在一种可能的设计中,该装置还包括:确定单元,该确定单元用于根据第一定位图片和第二定位图片的局部特征,确定第一定位图片和第二定位图片的第一相似度;局部特征包括以下中的一项或多项:定位图片中像素的灰度平均值,像素的灰度方差,像素的灰度协方差;该确定单元,还用于当第一相似度大于第一阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the apparatus further includes: a determining unit, configured to determine the first similarity between the first positioning picture and the second positioning picture according to the local features of the first positioning picture and the second positioning picture ; The local features include one or more of the following: the grayscale average value of the pixels in the positioning picture, the grayscale variance of the pixels, and the grayscale covariance of the pixels; the determining unit is also used for when the first similarity is greater than the first similarity. When a threshold is used, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
在一种可能的设计中,确定单元,还用于当第一相似度小于第一阈值时,确定第一定位图片和第二定位图片处于不同道路图层。In a possible design, the determining unit is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
在一种可能的设计中,确定单元,还用于根据第一定位图片和第二定位图片中对 应像素的相对高度信息,确定第一定位图片和第二定位图片的第二相似度;当第二相似度小于第二阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the determining unit is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture; When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
在一种可能的设计中,该确定单元,还用于当第二相似度大于第二阈值时,第一定位图片和第二定位图片处于不同道路图层。In a possible design, the determining unit is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
在一种可能的设计中,获取单元,具体用于针对第一定位图片和第二定位图片分别执行以下操作,以获取第一定位图片对应的特征指纹,和第二定位图片对应的特征指纹:删除定位图片中预设行和/或预设列的像素,以获取缩小的定位图片,根据缩小的定位图片中,各个像素的相对高度的均值,对缩小的定位图片进行归一化处理,根据归一化处理后的缩小的定位图片的各个像素值,确定定位图片对应的特征指纹;根据第一定位图片的特征指纹和第二定位图片的特征指纹,确定第一定位图片和第二定位图片的第二相似度,第二相似度为第一定位图片的特征指纹和第二定位图片的特征指纹的汉明距离。In a possible design, the acquisition unit is specifically configured to perform the following operations on the first positioning picture and the second positioning picture respectively, to obtain the feature fingerprint corresponding to the first positioning picture and the feature fingerprint corresponding to the second positioning picture: Delete the pixels in the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the average value of the relative heights of each pixel in the reduced positioning picture. Each pixel value of the reduced positioning picture after normalization is determined to determine the feature fingerprint corresponding to the positioning picture; according to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, the first positioning picture and the second positioning picture are determined. The second similarity is the Hamming distance between the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture.
第三方面,提供一种地图绘制装置,该地图绘制装置包括一个或多个处理器和一个或多个存储器;一个或多个存储器与所述一个或多个处理器耦合,所述一个或多个存储器存储有计算机指令;当所述一个或多个处理器执行所述计算机指令时,使得所述通信装置执行如第一方面及其可能的设计中任一项所述的地图绘制方法。In a third aspect, there is provided a mapping device comprising one or more processors and one or more memories; one or more memories coupled to the one or more processors, the one or more memories A memory stores computer instructions; when executed by the one or more processors, the computer instructions are caused to cause the communication device to perform the mapping method of any one of the first aspect and possible designs thereof.
示例性的,在调用存储器中的计算机指令时,该处理器用于获取第一待绘制区域的激光点云数据,第一待绘制区域包括处于不同平面的第一道路和第二道路;还用于根据激光点云数据,获取第一定位图片的集合和第二定位图片的集合;其中,任意两个第一定位图片的参考高度相差不超过第一阈值;任意两个第二定位图片的参考高度相差不超过第一阈值;第一定位图片和第二定位图片的参考高度差大于第二阈值;定位图片的参考高度是采集定位图片对应的激光点云数据时,对应设备所在路面的高度;第一阈值和第二阈值均为正数;处理器,用于将第一定位图片集合中的定位图片进行融合,获得第一道路对应的地图,将第二定位图片集合中的定位图片进行融合,获得第二道路对应的地图。Exemplarily, when invoking the computer instructions in the memory, the processor is configured to acquire the laser point cloud data of the first area to be drawn, where the first area to be drawn includes the first road and the second road on different planes; also used for According to the laser point cloud data, a set of first positioning pictures and a set of second positioning pictures are obtained; wherein, the reference height difference of any two first positioning pictures does not exceed the first threshold; the reference heights of any two second positioning pictures The difference does not exceed the first threshold; the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is the height of the road corresponding to the device when the laser point cloud data corresponding to the positioning picture is collected; The first threshold and the second threshold are both positive numbers; the processor is configured to fuse the positioning pictures in the first positioning picture set, obtain a map corresponding to the first road, and fuse the positioning pictures in the second positioning picture set, Obtain the map corresponding to the second road.
在一种可能的设计中,该激光点云数据包括:第一位置在全局坐标系下的三维坐标信息,其中,第一位置为采集激光点云数据的位置;该处理器,还用于根据第一位置的三维坐标信息,获取第一位置相对于采集激光点云数据时所行驶道路的相对高度信息。In a possible design, the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected; the processor is further configured to The three-dimensional coordinate information of the first position is obtained, and the relative height information of the first position relative to the road on which the laser point cloud data is collected is obtained.
在一种可能的设计中,该激光点云数据还包括:用于指示第一位置是否为车道的第一标识。In a possible design, the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
在一种可能的设计中,该处理器,用于根据激光点云数据,获取多个定位图片,每个定位图片对应的激光点云数据的采集时间在预设范围之内,每个定位图片中包括的像素的像素值由像素对应位置的相对高度信息和位置对应的激光点云数据的第一标识确定;该处理器,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,获取第一定位图片集合和第二定位图片集合。In a possible design, the processor is used to acquire multiple positioning pictures according to the laser point cloud data, the collection time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture The pixel value of the pixel included in the pixel value is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position; the processor is used for positioning the positioning pictures in the same road layer among the multiple positioning pictures Perform fusion to obtain the first positioning picture set and the second positioning picture set.
在一种可能的设计中,该装置还包括:处理器,该处理器用于确定第一定位图片和第二定位图片的相似度。该处理器,还用于根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,相似度用于指示第一定位图片和第二定位图片的相 似程度;其中,第一定位图片和第二定位图片为多个定位图片中,任意两个具有相同水平覆盖区域的定位图片。处理器,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,以获取处于第一道路对应的第一定位图片集合,以及第二道路对应的第二定位图片集合。In a possible design, the apparatus further includes: a processor configured to determine the similarity between the first positioning picture and the second positioning picture. The processor is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, and the similarity is used to indicate the similarity of the first positioning picture and the second positioning picture; wherein, the first positioning picture and the second positioning picture are similar. The positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures. The processor is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
在一种可能的设计中,该装置还包括:处理器,该处理器用于根据第一定位图片和第二定位图片的局部特征,确定第一定位图片和第二定位图片的第一相似度;局部特征包括以下中的一项或多项:定位图片中像素的灰度平均值,像素的灰度方差,像素的灰度协方差;该处理器,还用于当第一相似度大于第一阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the apparatus further includes: a processor, configured to determine the first similarity between the first positioning picture and the second positioning picture according to local features of the first positioning picture and the second positioning picture; The local features include one or more of the following: the grayscale average value of the pixels in the positioning image, the grayscale variance of the pixels, and the grayscale covariance of the pixels; the processor is also used for when the first similarity is greater than the first similarity When the threshold is set, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
在一种可能的设计中,处理器,还用于当第一相似度小于第一阈值时,确定第一定位图片和第二定位图片处于不同道路图层。In a possible design, the processor is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
在一种可能的设计中,处理器,还用于根据第一定位图片和第二定位图片中对应像素的相对高度信息,确定第一定位图片和第二定位图片的第二相似度;当第二相似度小于第二阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the processor is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture; When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
在一种可能的设计中,该处理器,还用于当第二相似度大于第二阈值时,第一定位图片和第二定位图片处于不同道路图层。In a possible design, the processor is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
在一种可能的设计中,处理器,具体用于针对第一定位图片和第二定位图片分别执行以下操作,以获取第一定位图片对应的特征指纹,和第二定位图片对应的特征指纹:删除定位图片中预设行和/或预设列的像素,以获取缩小的定位图片,根据缩小的定位图片中,各个像素的相对高度的均值,对缩小的定位图片进行归一化处理,根据归一化处理后的缩小的定位图片的各个像素值,确定定位图片对应的特征指纹;根据第一定位图片的特征指纹和第二定位图片的特征指纹,确定第一定位图片和第二定位图片的第二相似度,第二相似度为第一定位图片的特征指纹和第二定位图片的特征指纹的汉明距离。In a possible design, the processor is specifically configured to perform the following operations on the first positioning picture and the second positioning picture, respectively, to obtain a feature fingerprint corresponding to the first positioning picture and a feature fingerprint corresponding to the second positioning picture: Delete the pixels in the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the average value of the relative heights of each pixel in the reduced positioning picture. Each pixel value of the reduced positioning picture after normalization is determined to determine the feature fingerprint corresponding to the positioning picture; according to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, the first positioning picture and the second positioning picture are determined. The second similarity is the Hamming distance between the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture.
第四方面,提供一种芯片***,该芯片***可以应用于地图绘制装置中。示例性的,该芯片***包括接口电路和处理器;接口电路和处理器通过线路互联;接口电路用于从电子装置的存储器接收信号,并向处理器发送信号,信号包括存储器中存储的计算机指令;当处理器执行该计算机指令时,芯片***执行如第一方面及其可能的设计中任一项所述的地图绘制方法。In a fourth aspect, a chip system is provided, and the chip system can be applied to a map drawing device. Exemplarily, the chip system includes an interface circuit and a processor; the interface circuit and the processor are interconnected by a line; the interface circuit is used to receive a signal from a memory of the electronic device and send a signal to the processor, and the signal includes computer instructions stored in the memory. ; When the processor executes the computer instructions, the chip system executes the map drawing method as described in any one of the first aspect and its possible designs.
第五方面,提供一种计算机可读存储介质,该计算机可读存储介质包括计算机指令,当该计算机指令运行时,执行如第一方面及其可能的设计中任一项所述的地图绘制方法。In a fifth aspect, a computer-readable storage medium is provided, the computer-readable storage medium includes computer instructions that, when executed, perform the mapping method described in any one of the first aspect and possible designs thereof. .
第六方面,提供一种计算机程序产品,该计算机程序产品中包括指令,当该计算机程序产品在计算机上运行时,使得计算机可以根据该指令执行如第一方面及其可能的设计中任一项所述的地图绘制方法。A sixth aspect provides a computer program product, the computer program product includes instructions, when the computer program product runs on a computer, the computer can execute any one of the first aspect and its possible designs according to the instructions The described map drawing method.
应当理解的是,上述第二方面,第三方面,第四方面,第五方面以及第六方面提供的技术方案,其技术特征均可对应到第一方面及其可能的设计中提供的地图绘制方法,因此能够达到的有益效果类似,此处不再赘述。It should be understood that the technical features of the technical solutions provided in the second aspect, the third aspect, the fourth aspect, the fifth aspect and the sixth aspect can all correspond to the mapping provided in the first aspect and its possible designs. Therefore, the beneficial effects that can be achieved are similar, which will not be repeated here.
图1为一种通过激光点云的方法获取的高精度定位地图的示意图;Fig. 1 is a kind of schematic diagram of the high-precision positioning map obtained by the method of laser point cloud;
图2为一种道路场景的简化示意图;2 is a simplified schematic diagram of a road scene;
图3为本申请实施例提供的一种地图绘制方法的流程示意图;3 is a schematic flowchart of a map drawing method provided by an embodiment of the present application;
图4为本申请实施例提供的一种道路标识的示意图;4 is a schematic diagram of a road sign provided by an embodiment of the present application;
图5A为本申请实施例提供的一种相对高度的确定示意图;5A is a schematic diagram of determining a relative height according to an embodiment of the present application;
图5B为本申请实施例提供的又一种相对高度的确定示意图;FIG. 5B is a schematic diagram of yet another relative height determination provided by an embodiment of the present application;
图6为本申请实施例提供的一种定位图片的获取示意图;FIG. 6 is a schematic diagram of obtaining a positioning picture according to an embodiment of the present application;
图7为本申请实施例提供的另一种定位图片的获取示意图;7 is a schematic diagram of obtaining another positioning picture provided by an embodiment of the present application;
图8为本申请实施例提供的一种局部特征的获取方法的示意图;8 is a schematic diagram of a method for acquiring a local feature provided by an embodiment of the present application;
图9为本申请实施例提供的一种特征指纹的获取示意图;FIG. 9 is a schematic diagram of acquiring a characteristic fingerprint according to an embodiment of the present application;
图10为本申请实施例提供的一种基于特征指纹确定相似度的方法示意图;10 is a schematic diagram of a method for determining similarity based on feature fingerprints provided by an embodiment of the present application;
图11为本申请实施例提供的一组行为图片的对比示意图;11 is a schematic diagram of a comparison of a group of behavior pictures provided by an embodiment of the present application;
图12为本申请实施例提供的又一组定位图片的对比示意图;FIG. 12 is a comparative schematic diagram of another group of positioning pictures provided by an embodiment of the present application;
图13为本申请实施例提供的又一组定位图片的对比示意图;FIG. 13 is a comparative schematic diagram of another group of positioning pictures provided by an embodiment of the present application;
图14为本申请实施例提供的又一组定位图片的对比示意图;FIG. 14 is a schematic comparison diagram of another group of positioning pictures provided by an embodiment of the present application;
图15为本申请实施例提供的一种融合图片的示意图;15 is a schematic diagram of a fusion picture provided by an embodiment of the present application;
图16为本申请实施例提供的一种地图绘制装置的示意图;16 is a schematic diagram of a map drawing device provided by an embodiment of the present application;
图17为本申请实施例提供的又一种地图绘制装置的示意图;17 is a schematic diagram of another map drawing device provided by an embodiment of the present application;
图18为本申请实施例提供的一种芯片***的示意图。FIG. 18 is a schematic diagram of a chip system provided by an embodiment of the present application.
高精度定位地图是一种能够直接被设备所识别并使用的能够提供详细地图数据的地图。示例性的,根据该高精度定位地图,设备(如车辆)能够自行确定当前路况信息,并基于该路况信息实现自动驾驶。A high-precision positioning map is a map that can be directly recognized and used by the device and can provide detailed map data. Exemplarily, according to the high-precision positioning map, a device (such as a vehicle) can determine the current road condition information by itself, and implement automatic driving based on the road condition information.
目前,高精度定位地图可以通过图像和全球定位***(global posit ioning system,GPS)技术获取。在使用该方法获取高精度地图时,测量设备(如机器人)在需要绘制地图的区域的道路上行驶,通过拍摄等形式,以获取在道路上行驶过程中,在不同位置的环境图片。其中,不同的位置可以通过GPS定位获取。根据这些环境图片,并参考机器人所处位置,就可以获取对应的高精度定位地图。At present, high-precision positioning maps can be obtained through image and global positioning system (global positioning system, GPS) technology. When using this method to obtain a high-precision map, a measuring device (such as a robot) drives on the road in the area where the map needs to be drawn, and obtains environmental pictures at different locations during the process of driving on the road by taking pictures. Among them, different positions can be obtained through GPS positioning. According to these environmental pictures and referring to the position of the robot, the corresponding high-precision positioning map can be obtained.
然而,由于机器人拍摄获取的环境图片的精确度,以及GPS定位精确度的限制,使得基于该方法获取的高精度定位地图的精度相对较低。因此,根据图像和GPS技术绘制获取的高精度定位地图主要应用于自动驾驶等级在层2(layer 2,L2)或层3(layer3,L3)等对于地图精度要求较低的场景下。例如,该高精度定位地图可以用于支持高级驾驶辅助***(advanced driving assistance system)的实现。而在自动驾驶等级较高的场景(如自动驾驶等级在层4(layer 4,L4)或层5(layer 5,L5)的场景)中,由于对地图精度要求较高,因此无法使用通过上述图像和GPS技术绘制获取的高精度地图。However, due to the accuracy of the environment pictures captured by the robot and the limitation of GPS positioning accuracy, the accuracy of the high-precision positioning map obtained based on this method is relatively low. Therefore, the high-precision positioning map drawn and obtained according to the image and GPS technology is mainly used in scenarios with low requirements for map accuracy, such as automatic driving level at layer 2 (layer 2, L2) or layer 3 (layer 3, L3). For example, the hyperlocation map can be used to support the implementation of advanced driving assistance systems. However, in scenarios with a higher level of automatic driving (such as a scene where the level of automatic driving is at layer 4 (layer 4, L4) or layer 5 (layer 5, L5)), due to the high requirements for map accuracy, it cannot be used through the above Imagery and GPS technology to draw high-precision maps.
为了满足在自动驾驶等级较高的场景中对地图精度的要求,可以通过激光点云的测量方法获取具有较高精度的高精度定位地图。In order to meet the requirements for map accuracy in scenarios with high levels of automatic driving, a high-precision positioning map with high accuracy can be obtained through the measurement method of laser point cloud.
示例性的,多个不同的测量设备(如机器人)分别在需要绘制地图的道路上行驶, 通过设置在机器人上的激光测量模块等部件,获取在道路行驶过程中,不同位置对应的地图信息。其中,该地图信息可以包括通过激光测量模块测量得到的三维环境,回波强度等信息。在得到机器人在道路上行驶过程中获取的大量地图信息后,可以结合机器人在获取地图信息时的位姿(如机器人所处位置,以及机器人在获取该地图信息时的角度)以及对应的传感器信息,绘制对应的局部地图。作为一种实现方式,可以通过即时定位与地图构建(simultaneous localization and mapping,SLAM)的方法获取该局部地图。在获取局部地图后,可以通过对多个机器人采集的地图信息对应的多个局部地图进行融合,最终获取需要绘制地图的区域对应的高精度定位地图。需要说明的是,由于每个机器人的局部地图大小都是有限的,因此,为了能够准确地将不同机器人获取的局部地图融合在一起,需要保证不同机器人在测量过程中所覆盖的区域要有一定重叠区域。以便在进行地图融合时,可以根据不同局部地图中,用于标示同一位置(即重叠区域中的位置)信息的关系,确定两个局部地图之间的相互关系,进而进行准确的融合。Exemplarily, a plurality of different measuring devices (such as robots) are respectively driving on a road where a map needs to be drawn, and map information corresponding to different positions during road driving is obtained through components such as a laser measurement module arranged on the robot. Wherein, the map information may include information such as the three-dimensional environment measured by the laser measurement module, echo intensity, and the like. After obtaining a large amount of map information obtained by the robot while driving on the road, it can be combined with the robot's pose (such as the robot's position and the angle when the robot obtains the map information) and the corresponding sensor information when it obtains the map information. , draw the corresponding local map. As an implementation manner, the local map can be obtained by a method of simultaneous localization and mapping (SLAM). After the local map is obtained, a high-precision positioning map corresponding to the area where the map needs to be drawn can be finally obtained by fusing multiple local maps corresponding to the map information collected by multiple robots. It should be noted that since the size of the local map of each robot is limited, in order to accurately fuse the local maps obtained by different robots, it is necessary to ensure that the areas covered by different robots during the measurement process must have a certain amount. overlapping area. In order to perform map fusion, the relationship between the two local maps can be determined according to the relationship between the information used to indicate the same location (ie, the location in the overlapping area) in different local maps, and then accurate fusion can be performed.
另外,不同机器人在进行地图信息的采集时,可以是基于同一个坐标系下的,也可以是基于不同坐标系下的。在一些场景下,可以采用基于通用横墨卡托格网***(universal transverse mercator grid system,UTM)的通用坐标系作为所有机器人进行地图信息采集时的统一坐标系。在对使用该UTM坐标系获取的地图信息进行融合时,可以将不同机器人的位姿信息通过泰勒微分变换(或称为T变换)对应到该UTM坐标系中,以便根据机器人的位姿信息进行地图信息的融合。在另一些场景下,机器人的地图信息的采集也可以是根据不同坐标系进行的。在对具有不同坐标系的地图信息进行融合时,需要根据不同机器人采集到的地图信息中,重合部分的地图信息的差异,对不同地图信息进行归一化处理,以便实现对具有不同坐标系的地图信息进行顺利的融合。请参考图1,示出了一种通过激光点云的方法获取的高精度定位地图的示意图。如图1所示,高精度定位地图可通过俯视图的形式,显示对应区域内的水平面上的物体(如树木,房屋,道路等)的分布情况。在图1所示的高精度定位地图下,对于没有物体分布的区域对应位置的像素可以被显示为黑色。对于有物体分布的区域,则在对应位置的像素可以被显示为有灰度的颜色或白色。其中灰度可以地图数据中的回波参数等数据确定。In addition, when different robots collect map information, they may be based on the same coordinate system or based on different coordinate systems. In some scenarios, a universal coordinate system based on the universal transverse mercator grid system (UTM) can be used as the unified coordinate system for all robots to collect map information. When the map information obtained by using the UTM coordinate system is fused, the pose information of different robots can be mapped into the UTM coordinate system through Taylor differential transformation (or T transformation), so that the robot can carry out Fusion of map information. In other scenarios, the collection of the robot's map information may also be performed according to different coordinate systems. When the map information with different coordinate systems is fused, it is necessary to normalize the different map information according to the difference of the overlapped map information in the map information collected by different robots, so as to realize the integration of the map information with different coordinate systems. The map information is smoothly integrated. Please refer to FIG. 1 , which shows a schematic diagram of a high-precision positioning map obtained by a method of laser point cloud. As shown in Figure 1, the high-precision positioning map can display the distribution of objects (such as trees, houses, roads, etc.) on the horizontal plane in the corresponding area in the form of a top view. Under the high-precision positioning map shown in Figure 1, the pixels corresponding to the location of the area without object distribution can be displayed as black. For areas where objects are distributed, the pixels at the corresponding positions can be displayed in grayscale or white. The grayscale can be determined by data such as echo parameters in the map data.
基于上述说明,可以理解的是,在如图1所示的高精度地图中,由于法规中存在不能体现高度信息的要求,该地图中将具有三维空间分布的路况信息压缩为仅具有水平分布的路况,由此使得高度信息的丢失,进而导致车辆无法根据该地图区分不同高度下的路况分布。示例性的,请参考图2。如图2中的(a)所示,该场景中包括道路A,以及垂直道路高于道路A的道路B。通过目前技术中的激光点云的方法获取的高精度地图如图2中的(b)所示,其中,由于丢失了高度的相关信息,因此,基于该地图无法区分道路A和道路B的垂直分布关系,也就使得当有车辆行驶到道路A和道路B的投影交叉区域时,可能出现无法提供正确的定位信息等问题。特别是在车辆处于自动驾驶时,这种定位失败可能会导致车辆以及道路的损坏等严重问题。Based on the above description, it can be understood that, in the high-precision map as shown in Figure 1, due to the requirement that the height information cannot be reflected in the regulations, the road condition information with three-dimensional spatial distribution in the map is compressed into only the horizontal distribution. Therefore, the height information is lost, and the vehicle cannot distinguish the distribution of road conditions at different heights according to the map. For example, please refer to FIG. 2 . As shown in (a) of FIG. 2 , the scene includes a road A, and a road B whose vertical road is higher than the road A. The high-precision map obtained by the method of laser point cloud in the current technology is shown in (b) of Fig. 2, in which, since the relevant information of height is lost, the vertical direction of road A and road B cannot be distinguished based on this map. The distribution relationship means that when a vehicle travels to the projected intersection area of road A and road B, problems such as inability to provide correct positioning information may occur. Especially when the vehicle is in autonomous driving, this kind of localization failure can lead to serious problems such as damage to the vehicle as well as the road.
为了解决上述问题,本申请实施例提供一种地图绘制方法,能够对具有不同高度的道路分别绘制对应的定位地图,使得设备能够根据该多个高精度定位地图,并结合 自身所处位置,灵活选取对应高度下的高精度定位地图进行定位,能够有效避免由于高精度定位地图中不包括绝对高程信息导致的设备无法根据高精度定位地图获取准确的定位信息的问题。应当理解的是,由于目前的地图绘制方案中,存在上述无法区分不同高度的道路分布等问题,因此使得大范围的地图构建变得异常困难。而本申请实施例提供的地图绘制方法,能够有效地解决上述问题,因此能够支持大范围内的高精度定位地图的地图构建过程。In order to solve the above problems, the embodiments of the present application provide a map drawing method, which can draw corresponding positioning maps for roads with different heights, so that the device can flexibly locate the maps according to the multiple high-precision positioning maps and combine its own location. Selecting the high-precision positioning map at the corresponding height for positioning can effectively avoid the problem that the device cannot obtain accurate positioning information based on the high-precision positioning map due to the fact that the high-precision positioning map does not include absolute elevation information. It should be understood that, due to the above-mentioned problems such as the inability to distinguish the distribution of roads at different heights in the current map drawing solution, it becomes extremely difficult to construct a large-scale map. However, the map drawing method provided by the embodiment of the present application can effectively solve the above problem, and thus can support a map construction process of a high-precision positioning map in a large area.
示例性的,根据获取的激光点云数据可以获取不同道路对应的道路图层的定位图片,并根据这些定位图片的相似度,确定不同定位图片是否处于同一个道路对应的图层,以便对处于相同图层的定位图片进行融合,最终获取在不同高度下分布的道路的多个图层作为高精度定位地图。Exemplarily, according to the obtained laser point cloud data, the positioning pictures of the road layers corresponding to different roads can be obtained, and according to the similarity of these positioning pictures, it is determined whether the different positioning pictures are in the layer corresponding to the same road, so as to determine whether the different positioning pictures are in the layer corresponding to the same road. The positioning pictures of the same layer are fused, and finally multiple layers of roads distributed at different heights are obtained as a high-precision positioning map.
应当理解的是,本申请实施例提供的地图绘制方法,能够应用于高精度定位地图绘制的场景中,特别是能够高效地为自动驾驶所使用的高精度定位地图的绘制提供支持。当然,通过该方法所绘制获取的高精度定位地图也可应用于其他场景中,如可以应用于涉及智能机器人的自主移动定位等场景中。It should be understood that the map drawing method provided by the embodiments of the present application can be applied to the scene of high-precision positioning map drawing, and in particular, can efficiently provide support for the drawing of high-precision positioning maps used for automatic driving. Of course, the high-precision positioning map drawn and obtained by this method can also be applied to other scenarios, such as scenarios involving autonomous movement and positioning of intelligent robots.
以下结合附图对本申请实施例提供的技术方案进行详细说明。为了便于说明,其中以通过该地图绘制方法获取用于车辆进行自动驾驶的高精度定位地图为例。本申请实施例中,也可将高精度定位地图简称为定位地图。The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings. For the convenience of description, an example of obtaining a high-precision positioning map for automatic driving of a vehicle by the map drawing method is used. In this embodiment of the present application, the high-precision positioning map may also be referred to as a positioning map for short.
请参考如图3,为本申请实施例提供的一种地图绘制方法的流程示意图。如图3所示,该方法可以包括S301-S304。Please refer to FIG. 3 , which is a schematic flowchart of a map drawing method according to an embodiment of the present application. As shown in FIG. 3, the method may include S301-S304.
S301、获取待绘制区域的激光点云数据。S301. Acquire laser point cloud data of an area to be drawn.
在一些实现方式中,该激光点云数据可以通过机器人、车辆等测量设备获取。示例性的,下面以测量设备为设置有激光测量模块(如激光传感器)的车辆为例。车辆可以在需要绘制地图的区域内(如称为待绘制区域)的道路上行驶,并在行驶过程中的不同位置,采集与周围环境中的物体对应的激光点云数据。其中,车辆上的激光传感器可以发射出激光信号,并接收对应的反馈信号。根据反馈信号的相关参数获取即可获取对应的激光点云数据。例如,该激光信号对应的反馈信号对应的参数,可以包括该激光信号被反射位置(如称为激光点云对应的位置)在全局坐标系(如地理坐标系,笛卡尔坐标系等)下的XYZ三维坐标信息。其中,X和Y坐标可以用于标识激光点云对应的位置的水平坐标,Z坐标可以用于标识该激光点云对应的位置的垂直坐标。应当理解的是,该激光点云数据还可以包括其他信息,例如激光信号的回波次数,强度信息,类别,车辆的扫描角度,以及扫描方向等信息。In some implementations, the laser point cloud data can be acquired by measuring equipment such as robots and vehicles. Exemplarily, the following takes the measurement device as a vehicle provided with a laser measurement module (such as a laser sensor) as an example. The vehicle can drive on the road in the area that needs to be mapped (such as the area to be mapped), and collect laser point cloud data corresponding to objects in the surrounding environment at different positions during the driving process. Among them, the laser sensor on the vehicle can emit laser signals and receive corresponding feedback signals. The corresponding laser point cloud data can be obtained according to the relevant parameters of the feedback signal. For example, the parameters corresponding to the feedback signal corresponding to the laser signal may include the reflected position of the laser signal (such as the position corresponding to the laser point cloud) in the global coordinate system (such as geographic coordinate system, Cartesian coordinate system, etc.) XYZ three-dimensional coordinate information. The X and Y coordinates may be used to identify the horizontal coordinates of the position corresponding to the laser point cloud, and the Z coordinate may be used to identify the vertical coordinates of the position corresponding to the laser point cloud. It should be understood that the laser point cloud data may also include other information, such as the number of echoes of the laser signal, intensity information, category, scanning angle of the vehicle, and scanning direction and other information.
另外,本示例中,由于该定位地图用于自动驾驶的定位信息的获取,而车道标识相比于其他物体在车辆进行自动驾驶过程中更加重要,例如,通过道路标识能够准确地确定行驶道路的状况(如是否拐弯、前方是否有人行横道、车道数等)。因此,在本示例中,该激光信号对应的反馈信号对应的参数,还可以包括该激光信号对应位置是否为车道标识的标识。作为一种示例,该车道标识可以为车道线,路标等标识。为了便于说明,以下以车道标识为车道线为例。例如,如图4所示,在路面上,车道线即为如图4中所表示的车道标识。类似的,在如图4所示的匝道上以及高架上,也会有对应的车道标识。在结合上述说明,在该车道线对应位置采集获取的激光点云数据, 就可被打上是车道标识的标识。反之,对于其他非车道线对应位置采集获取的激光点云数据,就可以被打上非车道标识的标识。需要说明的是,在本申请的不同实现方式中,可以通过不同的方法达到上述示例中打上车道标识或者打上非车道标识的效果。例如,可以在需要打上车道标识的激光点云数据中的固定位(如激光点云数据对应的二进制数的固定帧位)置1(或置0),用于标示该激光点云对应位置是车道标识。又如,在需要打上非车道标识的激光点云数据中的固定位置0(或置1),或不做置空,用于标示该激光点云对应位置不是车道标识。在一些实施例中,该用于标识对应激光点云数据是否为车道标识的固定位也可以为激光点云数据中的约定标识。本申请实施例对于该车道标识的标识方法不做限定。为了便于说明,以下以车道线所在位置对应的激光点云数据用1标识,非车道线所在位置对应的激光点云数据用0标识为例。In addition, in this example, since the positioning map is used to obtain the positioning information of automatic driving, the lane marking is more important than other objects in the process of automatic driving of the vehicle. For example, the road marking can accurately determine the driving road. Conditions (such as whether to turn, whether there is a crosswalk ahead, number of lanes, etc.). Therefore, in this example, the parameter corresponding to the feedback signal corresponding to the laser signal may also include an indication of whether the position corresponding to the laser signal is a lane marking. As an example, the lane markings may be lane markings, road signs and other markings. For the convenience of description, the following takes the lane marking as the lane line as an example. For example, as shown in FIG. 4 , on the road surface, the lane line is the lane mark as shown in FIG. 4 . Similarly, there will also be corresponding lane markings on the ramp as shown in Figure 4 and on the elevated. In combination with the above description, the laser point cloud data collected at the corresponding position of the lane line can be marked as a lane mark. On the contrary, the laser point cloud data collected from the corresponding positions of other non-lane lines can be marked with non-lane markings. It should be noted that, in different implementation manners of the present application, the effect of marking a lane marking or marking a non-lane marking in the above example can be achieved by different methods. For example, a fixed bit in the laser point cloud data that needs to be marked with lane marks (such as the fixed frame position of the binary number corresponding to the laser point cloud data) can be set to 1 (or set to 0) to indicate that the corresponding position of the laser point cloud is Lane markings. For another example, the fixed position 0 (or 1) in the laser point cloud data that needs to be marked with a non-lane mark, or not blank, is used to indicate that the corresponding position of the laser point cloud is not a lane mark. In some embodiments, the fixed position for identifying whether the corresponding laser point cloud data is a lane mark may also be a convention mark in the laser point cloud data. This embodiment of the present application does not limit the marking method of the lane marking. For the convenience of description, the laser point cloud data corresponding to the position of the lane line is marked with 1, and the laser point cloud data corresponding to the position of the non-lane line is marked with 0 as an example.
需要说明的是,当该激光定位地图并非用于自动驾驶技术中,而是为其他场景提供定位和/或定位信息时,可以根据对应场景的需求,灵活选取对应的特征来代替本示例中的道路标识,以便能够更加准确地区分重要特征与其他特征,进而获取与该场景需求对应的高精度定位地图。It should be noted that when the laser positioning map is not used in automatic driving technology, but provides positioning and/or positioning information for other scenarios, the corresponding features can be flexibly selected according to the needs of the corresponding scenarios to replace the Road signs, in order to be able to more accurately distinguish important features from other features, and then obtain a high-precision positioning map corresponding to the needs of the scene.
另外,当待绘制区域较大,或者待绘制区域内的道路较多时,为了能够快速高效地进行激光点云数据的获取,可以设置多个车辆同步在不同的道路上采集激光点云数据,并确定每个激光点云对应位置是否为车道标识。在获取该激光点云数据后,车辆可以实时地将该激光点云数据存储到本地,或通过网络上传到云端(或服务器)。其中,本示例中的网络可以为第三代移动通信技术(3rd-Generation,3G),***移动通信技术(4th-Generation,4G),第五代移动通信技术(5th-Generat ion,5G)或其他能够用于进行数据传输的网络。车辆也可按预设的周期上传采集到的激光点云数据。本申请实施例对此不作限制。In addition, when the area to be drawn is large or there are many roads in the area to be drawn, in order to obtain laser point cloud data quickly and efficiently, multiple vehicles can be set up to collect laser point cloud data on different roads synchronously, and Determine whether the corresponding position of each laser point cloud is a lane marking. After acquiring the laser point cloud data, the vehicle can store the laser point cloud data locally in real time, or upload it to the cloud (or server) through the network. Wherein, the network in this example may be the third generation mobile communication technology (3rd-Generation, 3G), the fourth generation mobile communication technology (4th-Generation, 4G), the fifth generation mobile communication technology (5th-Generation, 5G) ) or other network that can be used for data transfer. The vehicle can also upload the collected laser point cloud data in a preset cycle. This embodiment of the present application does not limit this.
S302、对激光点云数据进行预处理。S302, preprocessing the laser point cloud data.
一般而言,由于车辆在采集激光点云数据时,存在空间干扰以及道路颠簸等因素,可能会导致车辆所收集到的激光点云数据不准确,由此使得所采集的激光点云数据无法被直接使用。因此,在本申请中,可以对一个或多个车辆采集获取的激光点云数据进行预处理,以便使得激光点云数据能够更加准确地表征待绘制区域中的物体的分布。示例性的,该预处理可以包括去噪、消旋、对齐以及降采样等处理操作中的一个或多个。Generally speaking, due to factors such as spatial interference and road bumps when the vehicle collects laser point cloud data, the laser point cloud data collected by the vehicle may be inaccurate, so that the collected laser point cloud data cannot be used. Use directly. Therefore, in the present application, the laser point cloud data acquired by one or more vehicles may be preprocessed, so that the laser point cloud data can more accurately characterize the distribution of objects in the area to be drawn. Exemplarily, the preprocessing may include one or more of processing operations such as denoising, derotation, alignment, and downsampling.
应当理解的是,本申请实施例提供的地图绘制方法,对处于不同高度层的道路图层分别进行绘制以获取对应高度层的定位地图。因此,在本示例中,可以在预处理的过程中,将激光点云数据中包括的基于全局坐标系的高度信息(如XYZ坐标中的Z坐标),转换为相对于采集该激光点云数据时所行驶的道路的相对高度,满足法律法规并以便于后续处理。示例性的,请参考图5A,为一种相对高度的确定示意图。如图5A所示的场景是车辆在地面上行驶的过程中进行的激光点云数据的采集。因此,可以地面为基准,确定该采集过程中获取的激光点云数据的相对高度。例如,匝道上的点A的相对高度为该点A到地面所在平面的距离H1,又如,高架上的点B的相对高度为该点B到地面所在平面的距离H2。It should be understood that, in the map drawing method provided by the embodiment of the present application, the road layers at different height layers are drawn respectively to obtain the positioning map of the corresponding height layer. Therefore, in this example, in the process of preprocessing, the height information based on the global coordinate system (such as the Z coordinate in the XYZ coordinates) included in the laser point cloud data can be converted into relative to the collected laser point cloud data. The relative height of the road being driven at the time, which satisfies laws and regulations and facilitates subsequent processing. Exemplarily, please refer to FIG. 5A , which is a schematic diagram of determining a relative height. The scene shown in Figure 5A is the collection of laser point cloud data during the process of the vehicle running on the ground. Therefore, the relative height of the laser point cloud data acquired in the acquisition process can be determined by taking the ground as a reference. For example, the relative height of point A on the ramp is the distance H1 from the point A to the plane on which the ground is located. For another example, the relative height of point B on the elevated is the distance H2 from the point B to the plane on which the ground is located.
需要说明的是,对于如图5A所示的匝道等高度并非固定的道路(如斜坡),可以 采用相对于道路边沿最近点的高度作为其相对于如图5A所示的地面的相对高度信息。类似的,当车辆行驶在匝道上时,则可以以匝道的路面为参考,确定不同物体的相对高度信息。It should be noted that, for a road (such as a slope) whose height is not fixed such as a ramp as shown in FIG. 5A , the height relative to the closest point of the road edge can be used as its relative height information relative to the ground as shown in FIG. 5A . Similarly, when the vehicle is driving on the ramp, the relative height information of different objects can be determined with reference to the road surface of the ramp.
示例性的,图5B示出了在XOZ平面下的相对高度的确定示意图。以车辆行驶在地面上进行数据采集为例。如图中的高架与地面为平行的两条道路,因此,高架对应的激光点云数据的相对高度均为H3。例如,高架对应的激光点云数据中包括如图5B所示的P3,其相对高度为H3。其中,该相对高度H3可以根据采集获取的路面的绝对高度(如H
路面)与高架的绝对高度(如H
高架)确定。例如,H3=H
高架-H
路面。
Exemplarily, FIG. 5B shows a schematic diagram of the determination of the relative height under the XOZ plane. Take the vehicle driving on the ground for data collection as an example. As shown in the figure, the elevated and the ground are two parallel roads. Therefore, the relative height of the laser point cloud data corresponding to the elevated is H3. For example, the laser point cloud data corresponding to the overhead includes P3 as shown in Figure 5B, and its relative height is H3. Wherein, the relative height H3 may be determined according to the absolute height of the road (eg, H road ) and the absolute height of the elevated (eg, H elevated ) acquired through collection. For example, H3 = H elevated - H road .
匝道由于需要与地面以及高架接驳,因此,匝道上不同激光点云数据相对地面的高度可能是不同的。如图5B所示,匝道上P4点的相对高度可以为H4,匝道上P5点的相对高度可以为H5。其中,该相对高度H4可以根据P4点的绝对高度H
P4和H
路面确定,例如,H4=H
P4-H
路面。类似的,该相对高度H5可以根据P5点的绝对高度H
P5和H
路面确定,例如,H5=H
P5-H
路面。
Since the ramp needs to be connected to the ground and elevated, the heights of different laser point cloud data on the ramp relative to the ground may be different. As shown in FIG. 5B , the relative height of point P4 on the ramp may be H4, and the relative height of point P5 on the ramp may be H5. Wherein, the relative height H4 can be determined according to the absolute height H P4 of point P4 and the H road surface , for example, H4=H P4 -H road surface . Similarly, the relative height H5 can be determined according to the absolute height H P5 of point P5 and the H road surface , for example, H5=H P5 -H road surface .
S303、根据激光点云数据,获取多个定位图片。S303. Acquire multiple positioning pictures according to the laser point cloud data.
本申请实施例中,可以对一个或多个车辆采集的激光点云数据进行处理,以获取多个定位图片。其中,每个定位图片都可以对应到XOY平面上一个道路图层中的一部分图像。示例性的,可以根据同一辆车在连续的一段时间(如T1)内所获取激光点云数据确定对应的定位图片。由此保证该定位图标片中不会同时包括处于不同都图层的信息。其中,该T1的长短,取决于想要获取的定位图片的尺寸。例如,以每个定位图片的尺寸为100*100米为例,可以根据车辆的行驶速度等信息确定T1的长短,并选取该T1时间内的激光点云数据进行融合以获取对应的定位图片。需要说明的是,在本申请实施例中,多个定位图片中任意两个定位图片的尺寸可以相同,也可不同。以下不同定位图片的尺寸相同为例进行说明。In this embodiment of the present application, the laser point cloud data collected by one or more vehicles may be processed to obtain multiple positioning pictures. Among them, each positioning image can correspond to a part of the image in a road layer on the XOY plane. Exemplarily, the corresponding positioning picture can be determined according to the laser point cloud data obtained by the same vehicle in a continuous period of time (eg T1). Therefore, it is ensured that the positioning icon piece does not include information in different layers at the same time. The length of the T1 depends on the size of the positioning image to be obtained. For example, taking the size of each positioning picture as 100*100 meters as an example, the length of T1 can be determined according to information such as the driving speed of the vehicle, and the laser point cloud data within the T1 period can be selected for fusion to obtain the corresponding positioning picture. It should be noted that, in this embodiment of the present application, the sizes of any two positioning pictures among the multiple positioning pictures may be the same or different. The following is an example of the same size of different positioning images.
另外,为了能够保证定位图片不会由于其中包括的信息的高度差过大而产生畸变等问题,本申请实施例的一些实现方式中,可以在根据激光点云数据获取定位图片时,限定在一个定位图片中不包括高度差大于一定阈值的激光点云数据。例如,可以限定一个定位图片中不包括高度差大于4米的激光点云数据。In addition, in order to ensure that the positioning picture will not be distorted due to the excessive height difference of the information contained therein, in some implementations of the embodiments of the present application, when obtaining the positioning picture according to the laser point cloud data, the positioning picture may be limited to one The laser point cloud data whose height difference is greater than a certain threshold are not included in the positioning image. For example, it can be defined that a positioning image does not include laser point cloud data with a height difference greater than 4 meters.
在本申请中,在根据激光点云数据获取定位图片时,定位图片中的每个像素可以对应到一组具有相同水平坐标(即X坐标以及Y坐标)但具有不同相对高度的激光点云数据。应当理解的是,定位图片可以为灰度平面地图,每个像素的灰度可以根据与之对应的激光点云数据确定。需要说明的是,在一些实现方式中,定位图片中的每个像素还可以是对应到一组具有形同水平坐标范围的激光点云数据。例如,每个像素可以对应的激光点云数据可以为X坐标在[X1,X2]范围内的,且Y坐标在[Y1,Y2]范围内的所有激光点云数据。这样,就可以通过灵活设置[X1,X2]和/或[Y1,Y2]的取值范围,使得定位图片中每个像素能够表征更多激光点云数据对应的信息,进而达到精简定位图片数量的目的。以下以一个像素对应一组水平坐标(即X1=X2,Y1=Y2)下不同高度的激光点云数据为例。In this application, when the positioning picture is obtained according to the laser point cloud data, each pixel in the positioning picture can correspond to a set of laser point cloud data with the same horizontal coordinates (ie X coordinates and Y coordinates) but with different relative heights . It should be understood that the positioning picture can be a grayscale plane map, and the grayscale of each pixel can be determined according to the corresponding laser point cloud data. It should be noted that, in some implementation manners, each pixel in the positioning picture may also correspond to a set of laser point cloud data having the same horizontal coordinate range. For example, the laser point cloud data corresponding to each pixel may be all laser point cloud data whose X coordinate is in the range of [X1, X2] and the Y coordinate is in the range of [Y1, Y2]. In this way, by flexibly setting the value range of [X1, X2] and/or [Y1, Y2], each pixel in the positioning picture can represent more information corresponding to the laser point cloud data, thereby reducing the number of positioning pictures the goal of. In the following, one pixel corresponds to laser point cloud data of different heights under a set of horizontal coordinates (ie X1=X2, Y1=Y2) as an example.
可以理解的是,由于定位图片为XOY平面内的二维图像,因此,在该定位图片上可以直观地展示出XOY平面上对应道路图层的分布情况。在本申请实施例中,还可以 通过不同像素的灰度信息,标识对应位置的相对高度信息以及道路标示等信息。It can be understood that, since the positioning picture is a two-dimensional image in the XOY plane, the distribution of the corresponding road layer on the XOY plane can be visually displayed on the positioning picture. In this embodiment of the present application, information such as the relative height information of the corresponding position and the road sign can also be identified by using the grayscale information of different pixels.
示例性的,每个像素都的灰度信息都可以对应到一个单通道多比特的二进制数。该二进制数的不同比特位可以用于标识该像素对应位置的物体的相对高度等信息。将该二进制数换算为十进制后即可对应到该像素对应的灰度。Exemplarily, the grayscale information of each pixel may correspond to a single-channel multi-bit binary number. Different bits of the binary number can be used to identify information such as the relative height of the object at the corresponding position of the pixel. After converting the binary number to decimal, it can correspond to the grayscale corresponding to the pixel.
以下对该单通道多比特的二进制数的确定方法进行举例说明。以每个像素的灰度信息对应于单通道8比特的二进制数为例。在该示例中,该8比特的二进制数换算成十进制,该对应的十进制数即可为对应像素的灰度。在该8比特中,可以采用7个比特位(如第0位到第6位)标识一个水平坐标下激光点云的高度分布情况,采用剩余的1个比特位(如第7位)标识对应位置的物体是否为车道标识。在一些实现方式中,该8比特的二进制数的第0位到第6位的每个比特位可以根据如下表1所示对应关系进行填充。The method for determining the single-channel multi-bit binary number is exemplified below. Take the grayscale information of each pixel corresponding to a single-channel 8-bit binary number as an example. In this example, the 8-bit binary number is converted into decimal, and the corresponding decimal number can be the grayscale of the corresponding pixel. Among the 8 bits, 7 bits (such as the 0th to the 6th position) can be used to identify the height distribution of the laser point cloud in a horizontal coordinate, and the remaining 1 bit (such as the 7th position) can be used to identify the corresponding Whether the object at the location is a lane marker. In some implementations, each bit of the 0th bit to the 6th bit of the 8-bit binary number may be filled according to the corresponding relationship shown in Table 1 below.
表1Table 1
比特位 |
00 | 11 | 22 | 33 | 44 | 55 | 66 |
高度特征 |
0或10 or 1 | 0或10 or 1 | 0或10 or 1 | 0或10 or 1 | 0或10 or 1 | 0或10 or 1 | 0或10 or 1 |
对应高度关系/米Corresponding height relation/m | 0.5-10.5-1 | 1-1.51-1.5 | 1.5-21.5-2 | 2-2.52-2.5 | 2.5-32.5-3 | 3-43-4 | 4-54-5 |
根据上述如表1所示的对应关系,如果在对应水平坐标下,相对高度为0.5-1米的区域内存在物体,则将第0位填充为1,反之,如果对应水平坐标下,相对高度为0.5-1米的区域内不存在位图,则将第0位填充为0。类似的,如果在对应水平坐标下,相对高度为1-1.5米的区域内存在物体,则将第1位填充为1,反之,如果对应水平坐标下,相对高度为1-1.5米的区域内不存在位图,则将第1位填充为0。如果在对应水平坐标下,相对高度为1.5-2米的区域内存在物体,则将第2位填充为1,反之,如果对应水平坐标下,相对高度为1.5-2米的区域内不存在位图,则将第2位填充为0。如果在对应水平坐标下,相对高度为2-2.5米的区域内存在物体,则将第3位填充为1,反之,如果对应水平坐标下,相对高度为2-2.5米的区域内不存在位图,则将第3位填充为0。如果在对应水平坐标下,相对高度为2.5-3米的区域内存在物体,则将第4位填充为1,反之,如果对应水平坐标下,相对高度为2.5-3米的区域内不存在位图,则将第4位填充为0。如果在对应水平坐标下,相对高度为3-4米的区域内存在物体,则将第5位填充为1,反之,如果对应水平坐标下,相对高度为3-4米的区域内不存在位图,则将第5位填充为0。如果在对应水平坐标下,相对高度为4-5米的区域内存在物体,则将第6位填充为1,反之,如果对应水平坐标下,相对高度为4-5米的区域内不存在位图,则将第6位填充为0。According to the above correspondence shown in Table 1, if there is an object in the area with a relative height of 0.5-1 meters under the corresponding horizontal coordinate, fill the 0th bit with 1, otherwise, if the corresponding horizontal coordinate, the relative height If there is no bitmap in the area of 0.5-1 meters, the 0th bit is filled with 0. Similarly, if there is an object in the area with a relative height of 1-1.5 meters under the corresponding horizontal coordinates, fill the first digit with 1. On the contrary, if the corresponding horizontal coordinates are in the area with a relative height of 1-1.5 meters If there is no bitmap, the first bit is filled with 0. If there is an object in the area with a relative height of 1.5-2 meters under the corresponding horizontal coordinate, fill the second bit with 1. On the contrary, if there is no object in the area with a relative height of 1.5-2 meters under the corresponding horizontal coordinate Figure, then fill the second bit with 0. If there is an object in the area with a relative height of 2-2.5 meters under the corresponding horizontal coordinate, fill the third bit with 1, otherwise, if there is no object in the area with a relative height of 2-2.5 meters under the corresponding horizontal coordinate Figure, then fill the 3rd bit with 0. If there is an object in the area with a relative height of 2.5-3 meters under the corresponding horizontal coordinates, fill the fourth bit with 1. On the contrary, if there is no object in the area with a relative height of 2.5-3 meters under the corresponding horizontal coordinates Figure, then fill the 4th bit with 0. If there is an object in the area with a relative height of 3-4 meters under the corresponding horizontal coordinate, fill the 5th bit with 1. On the contrary, if there is no object in the area with a relative height of 3-4 meters under the corresponding horizontal coordinate Figure, then fill the 5th bit with 0. If there is an object in the area with a relative height of 4-5 meters under the corresponding horizontal coordinate, fill the sixth bit with 1. On the contrary, if there is no object in the area with a relative height of 4-5 meters under the corresponding horizontal coordinate Figure, then fill the 6th bit with 0.
需要说明的是,上述表1示出了仅为一种对应关系的示例,在另一些实现方式中,也可采用更多或更好的比特位对相同水平坐标中的多个激光点云的高度分布进行标识。其每个比特位所对应的相对高度信息也可不同,本申请实施例对此不作限制。It should be noted that the above Table 1 shows only an example of a corresponding relationship. In other implementations, more or better bits may also be used to match multiple laser point clouds in the same horizontal coordinate. height distribution for identification. The relative height information corresponding to each bit may also be different, which is not limited in this embodiment of the present application.
在上述表1的基础上,可以利用该8比特二进制数中的剩余一位(如第7位)来标识该水平坐标下的激光点云是否是车道标识。例如,当该激光点云是车道标识时,则将该第7位设置为1。对应的,当该激光点云不是车道标识时,则将该第7位设置为0。On the basis of the above Table 1, the remaining one bit (eg, the seventh bit) in the 8-bit binary number can be used to identify whether the laser point cloud under the horizontal coordinate is a lane mark. For example, when the laser point cloud is a lane marking, set the 7th bit to 1. Correspondingly, when the laser point cloud is not a lane mark, the seventh bit is set to 0.
这样,就可以通过8比特二进制数确定具有相同水平坐标的多个激光点云的垂直 分布情况。接着可以根据该8比特二进制数确定该水平坐标下对应定位图片的像素的灰度。例如,可以将该8比特二进制数转换为十进制数,作为该像素的灰度,并以该灰度填充像素,最终获取定位图片。In this way, the vertical distribution of multiple laser point clouds with the same horizontal coordinates can be determined by 8-bit binary numbers. Then, the grayscale of the pixel corresponding to the positioning picture under the horizontal coordinate can be determined according to the 8-bit binary number. For example, the 8-bit binary number can be converted into a decimal number as the grayscale of the pixel, and the pixel can be filled with the grayscale, and finally a positioning picture can be obtained.
作为一种示例,请参考图6。其中继续以通过单通道8比特的二进制数标识高度分布情况为例。如图6中的(a)所示,在该水平坐标下包括一棵树,可以通过上述方法中的说明,对第0位到第6位进行填充,如可以填充获取00011111的二进制数用于标识该水平坐标下,高度的分布情况。另外,如图6中的(a)所示,该二进制数的第7位可以用于标识该水平坐标下的激光点云是否为车道标识。由于图中的激光点云的物体为一棵树,因此,可以将该第7位置0。由此,即可获取包括高度分布以及是否为车道标识的信息的8比特二进制数。如图6中的(b)所示,将该8比特二进制数转换为十进制数(如00011111对应31),即可获取该水平坐标下对应像素的灰度。类似的,可以根据该方法确定定位图片中,其他像素点的灰度,进行填充,即可获取对应的定位图片。应当理解的是,一般而言,水平坐标下的激光点云对应与车道标识时,则该水平坐标就对应与道路上的一个点,因此其上不会有其他垂直分布。示例性的,如图7中的(a)所示,水平坐标为(X1,Y1)的位置对应一棵树,因此其垂直分布可以通过如图7中的(a)所示的00111111来标识。水平坐标为(X2,Y2)的位置对应道路上的车道线,其垂直分布可以通过如图7中的(a)所示的10000000来标识。由此,即可获取如图7中的(b)所示的定位图片。As an example, please refer to FIG. 6 . Continuing to take the case of identifying the height distribution by a single-channel 8-bit binary number as an example. As shown in (a) of Figure 6, a tree is included in the horizontal coordinate, and the 0th to 6th bits can be filled by the description in the above method. For example, the binary number of 00011111 can be filled and used for Identifies the distribution of heights under this horizontal coordinate. In addition, as shown in (a) of FIG. 6 , the seventh bit of the binary number can be used to identify whether the laser point cloud under the horizontal coordinate is a lane mark. Since the object of the laser point cloud in the figure is a tree, the seventh position can be 0. Thereby, an 8-bit binary number including information on the height distribution and whether it is a lane marking can be obtained. As shown in (b) of FIG. 6 , by converting the 8-bit binary number into a decimal number (for example, 00011111 corresponds to 31), the grayscale of the corresponding pixel under the horizontal coordinate can be obtained. Similarly, the grayscales of other pixels in the positioning picture can be determined according to this method, and the corresponding positioning pictures can be obtained by filling them. It should be understood that, generally speaking, when the laser point cloud in the horizontal coordinate corresponds to the lane marking, the horizontal coordinate corresponds to a point on the road, so there will be no other vertical distribution on it. Exemplarily, as shown in (a) in FIG. 7 , the position whose horizontal coordinates are (X1, Y1) corresponds to a tree, so its vertical distribution can be identified by 00111111 as shown in (a) in FIG. 7 . . The position whose horizontal coordinate is (X2, Y2) corresponds to the lane line on the road, and its vertical distribution can be identified by 10000000 as shown in (a) of FIG. 7 . Thereby, the positioning picture as shown in (b) in FIG. 7 can be obtained.
应当理解的是,以上示例中,是以定位地图以及定位图片均为灰度图为例进行说明的。在另一些实施例中,如果要绘制的定位地图以及定位图片为彩色图(如RGB图)时,则每个像素的RGB色谱也可对应到一个单通道的多比特二进制数,具体的执行方法与上述灰度图的绘制方法类似,此处不再赘述。It should be understood that, in the above example, the positioning map and the positioning picture are both grayscale images as an example for description. In other embodiments, if the positioning map to be drawn and the positioning picture are color images (such as RGB images), the RGB color spectrum of each pixel can also correspond to a single-channel multi-bit binary number. The specific execution method It is similar to the above-mentioned drawing method of the grayscale image, and will not be repeated here.
S304、根据多个定位图片,获取不同道路图层对应的定位地图。S304 , according to the multiple positioning pictures, obtain positioning maps corresponding to different road layers.
应当理解的是,将这多个定位图片进行融合,即可获取定位地图。本申请实施例中,可以将不同道路图层的定位图片分别进行融合,以获取对应道路图层的定位地图。需要说明的是,在上述说明中的道路图层可以对应到具有不同高度的道路。也就是说,本申请实施例中,可以针对具有不同高度的道路,确定与该道路的道路图层所对应的定位图片集合。例如,第一道路可以对应于第一道路图层,该第一道路图层可以包括上述多个定位图片中,车辆行驶在第一道路上时,采集获取的激光点云数据所构成的定位图片的集合。类似的,第二道路可以对应于第二道路图层,该第二道路图层可以包括上述多个定位图片中,车辆行驶在第二道路上时,采集获取的激光点云数据所构成的定位图片的集合。在一些实现方式中,可以第一道路的绝对高度(即第一道路在全局坐标系下的高度)称为车辆在该第一道路上行驶过程中,采集获取的激光点云数据所对应的定位图片的参考高度。类似的,可以第二道路的绝对高度(即第二道路在全局坐标系下的高度)称为车辆在该第二道路上行驶过程中,采集获取的激光点云数据所对应的定位图片的参考高度。因此,对于相同道路图层中包括的任意两个定位图片的参考高度都可以包括在一个预设的范围内。也就是说,对于相同道路图层中包括的任意两个定位图片的参考高度差不超过第一阈值。相对的,对于不同道路图层中包括的定位图片的参考高度可以不在预设的范围之内。也就是说,对于不同道路图层中 包括的任意两个定位图片的参考高度差大于第二阈值。第一阈值可以与第二阈值相同,也可以与第二阈值不同。其具体的情况可以根据实际需求灵活选取或设置,本申请实施例对此不作限制。It should be understood that the positioning map can be obtained by fusing the multiple positioning pictures. In the embodiment of the present application, the positioning pictures of different road layers may be fused respectively to obtain the positioning map of the corresponding road layer. It should be noted that the road layers in the above description may correspond to roads with different heights. That is to say, in this embodiment of the present application, for a road with different heights, a set of positioning pictures corresponding to the road layer of the road may be determined. For example, the first road may correspond to the first road layer, and the first road layer may include the positioning pictures formed by the collected laser point cloud data when the vehicle drives on the first road among the above-mentioned multiple positioning pictures. collection. Similarly, the second road may correspond to the second road layer, and the second road layer may include the positioning formed by the collected laser point cloud data when the vehicle is driving on the second road in the above-mentioned multiple positioning pictures. Collection of pictures. In some implementations, the absolute height of the first road (that is, the height of the first road in the global coordinate system) may be referred to as the location corresponding to the laser point cloud data collected and acquired during the vehicle traveling on the first road The reference height of the image. Similarly, the absolute height of the second road (that is, the height of the second road in the global coordinate system) can be referred to as the reference of the positioning picture corresponding to the collected laser point cloud data when the vehicle is traveling on the second road. high. Therefore, the reference heights for any two positioning pictures included in the same road layer can be included in a preset range. That is, the reference height difference for any two positioning pictures included in the same road layer does not exceed the first threshold. On the contrary, the reference heights for the positioning pictures included in different road layers may not be within the preset range. That is to say, the reference height difference for any two positioning pictures included in different road layers is greater than the second threshold. The first threshold may be the same as the second threshold, or may be different from the second threshold. The specific situation can be flexibly selected or set according to actual needs, which is not limited in this embodiment of the present application.
在本申请实施例中,为了确定不同的定位图片对应的道路图层,可以确定不同定位图片之间是否处于相同的道路图层。如果定位图片处于不同的道路图层,则不需要对对应的定位图片进行融合。In this embodiment of the present application, in order to determine the road layers corresponding to different positioning pictures, it may be determined whether different positioning pictures are in the same road layer. If the positioning pictures are in different road layers, the corresponding positioning pictures do not need to be fused.
本申请实施例中,可以通过判断覆盖相同水平区域的两个或多个定位图片的相似度,确定定位图片是否位于同一个道路图层。可以理解的是,当所有定位图片所覆盖的水平区域不同时,则认为当前待绘制区域只存在一个道路图层,因此直接对着多个定位图片进行融合即可获取道路该待绘制区域的定位地图。In the embodiment of the present application, it can be determined whether the positioning pictures are located in the same road layer by judging the similarity of two or more positioning pictures covering the same horizontal area. It can be understood that when the horizontal areas covered by all the positioning pictures are different, it is considered that there is only one road layer in the current area to be drawn, so the location of the road to be drawn can be obtained by directly merging multiple positioning pictures. map.
以下结合示例,对判断覆盖相同水平区域的多个定位图片是否处于相同道路图层的方法进行说明。其中,以定位图片1和定位图片2覆盖了相同的水平区域为例。The method for judging whether multiple positioning pictures covering the same horizontal area are in the same road layer is described below with reference to an example. The positioning picture 1 and the positioning picture 2 cover the same horizontal area as an example.
示例性的,在一些实施例中,可以通过局部纹理及全局参考法,确定两个定位图片(如定位图片1和定位图片2)是否处于相同道路图层。在另一些实施例中,可以通过地图特征指纹法确定定位图片1和定位图片2是否处于相同道路图层。以下对这两种方法分别进行说明。Exemplarily, in some embodiments, it can be determined whether two positioning pictures (eg, positioning picture 1 and positioning picture 2 ) are in the same road layer through local texture and global reference method. In other embodiments, it may be determined whether the positioning picture 1 and the positioning picture 2 are in the same road layer through the map feature fingerprinting method. The two methods are described below.
1、局部纹理及全局参考法。1. Local texture and global reference method.
作为一种示例,针对定位图片1和定位图片2,可以分别采用窗口滑动的方法,确定每个定位图片中,局部特征的均值,和/或方差,和/或协方差。根据两个定位图片的局部特征,确定两个定位图片的相似度。进而根据该相似度与预设的阈值(如阈值1)的大小关系,确定该两个定位图片是否处于相同的道路图层。例如,当该相似度大于阈值1时,则认为两个定位图片处于相同的道路图层。反之,当该相似度小于阈值1时,则认为两个定位图片处于不同的道路图层。以下以局部特征包括均值,方差以及协方差为例。As an example, for positioning picture 1 and positioning picture 2, a window sliding method may be used respectively to determine the mean, and/or variance, and/or covariance of local features in each positioning picture. According to the local features of the two positioning pictures, the similarity of the two positioning pictures is determined. Further, according to the magnitude relationship between the similarity and a preset threshold (eg, threshold 1), it is determined whether the two positioning pictures are in the same road layer. For example, when the similarity is greater than the threshold 1, it is considered that the two positioning pictures are in the same road layer. Conversely, when the similarity is less than the threshold 1, it is considered that the two positioning pictures are in different road layers. The following takes local features including mean, variance and covariance as an example.
以下结合图8进行说明。图8为本申请实施例提供的一种局部特征的获取方法的示意图。其中,以定位图片包括6*6个像素,窗口的尺寸为3*3,滑动步长为1为例。如图8中的(a)所示,该窗口的初始位置可以位于定位图片的左上角,对应覆盖定位图片中左上角三行三列的像素。可以根据该窗口中9个像素的灰度,确定该窗口的灰度平均μ
1,灰度的方差σ
1以及灰度的协方差。在获取了该位置的局部特征后,可以将该窗口向右滑动1个像素到如图8中的(b)所示的位置。并获取该位置的灰度平均μ
2,灰度的方差σ
2以及灰度的协方差。此后,还可继续向右滑动,以获取其他位置的局部特征。在向右滑动到无法继续滑动时,可以将该窗口向下滑动一个像素(如移动到如图8中的(c)所示的位置),并参考上述方法获取其他位置的局部特征。如此重复,直至将窗口移动到如图8中的(d)所示的位置,并获取该位置的局部特征。这样,就获取了该定位图片对应的多个局部特征。例如,如图8所示的可以获取每个窗口位置对应的共16组局部特征。每组局部特征中均包括对应位置的灰度平均,灰度的方差以及灰度的协方差。
The following description will be made with reference to FIG. 8 . FIG. 8 is a schematic diagram of a method for acquiring a local feature provided by an embodiment of the present application. The positioning image includes 6*6 pixels, the size of the window is 3*3, and the sliding step size is 1 as an example. As shown in (a) of FIG. 8 , the initial position of the window may be located at the upper left corner of the positioning picture, corresponding to pixels covering three rows and three columns in the upper left corner of the positioning picture. According to the gray levels of the 9 pixels in the window, the gray level average μ 1 of the window, the gray level variance σ 1 and the gray level covariance can be determined. After obtaining the local features of the position, the window can be slid to the right by 1 pixel to the position shown in (b) of FIG. 8 . And obtain the gray average μ 2 , the gray variance σ 2 and the gray covariance of the position. After that, you can continue to swipe right to get local features in other locations. When swiping to the right to the point where it cannot continue to swipe, the window can be slid down by one pixel (for example, moved to the position shown in (c) in Figure 8), and the local features of other positions can be obtained by referring to the above method. This is repeated until the window is moved to the position shown in (d) in FIG. 8 , and the local features of the position are acquired. In this way, multiple local features corresponding to the positioning image are acquired. For example, as shown in Figure 8, a total of 16 sets of local features corresponding to each window position can be obtained. Each group of local features includes the average gray level of the corresponding position, the variance of the gray level, and the covariance of the gray level.
在获取了上述局部特征后,可以分别针对灰度平均,灰度的方差,以及灰度的协方差,构建两个定位图片的对比函数。After the above-mentioned local features are obtained, a contrast function of the two positioning images can be constructed for the average gray level, the variance of the gray level, and the covariance of the gray level, respectively.
示例性的,灰度平均(或称为特征均值)的对比函数可以通过如下公式(1)进行评估。Exemplarily, the contrast function of the gray-scale average (or referred to as the feature mean) can be evaluated by the following formula (1).
其中,l(x,y)为窗口处于某一位置时,与该位置对应的定位图片1和定位图片2之间的特征均值对比值。μ
x为该位置对应的定位图片1的特征均值。μ
y为该位置对应的定位图片2的特征均值。C
1为常数。作为一种示例,μ
x可以根据如下公式获取:
其中,H为窗口的高度,W为窗口的宽度,X(i,j)为像素为(i,j)对应位置处的像素值(如灰度)。
Among them, l(x, y) is the feature mean comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position. μx is the feature mean of the positioning picture 1 corresponding to the position. μ y is the feature mean of the positioning picture 2 corresponding to the position. C 1 is a constant. As an example, μ x can be obtained according to the following formula: Among them, H is the height of the window, W is the width of the window, and X(i,j) is the pixel value (such as grayscale) at the corresponding position of the pixel (i,j).
灰度的方差(或称为特征方差)的对比函数可以通过如下公式(2)进行评估。The contrast function of the variance of the gray level (or called the characteristic variance) can be evaluated by the following formula (2).
其中,c(x,y)为窗口处于某一位置时,与该位置对应的定位图片1和定位图片2之间的特征方差对比值。σ
x为该位置对应的定位图片1的特征方差。σ
y为该位置对应的定位图片2的特征方差。C
2为常数。作为一种示例,σ
x可以根据如下公式获取:
Among them, c(x, y) is the feature variance comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position. σ x is the feature variance of the positioning picture 1 corresponding to the position. σ y is the feature variance of the positioning picture 2 corresponding to the position. C2 is a constant. As an example, σ x can be obtained according to the following formula:
灰度的协方差(或称为特征协方差)的对比函数可以通过如下公式(3)进行评估。The contrast function of the covariance of gray levels (or called feature covariance) can be evaluated by the following formula (3).
其中,s(x,y)为窗口处于某一位置时,与该位置对应的定位图片1和定位图片2之间的特征协方差对比值。σ
xy为定位图片1中的对应像素(如X(i,j)),和定位图片2中的对应像素(如Y(i,j))的协方差。作为一种示例,该协方差可以通过如下公式获取:
Among them, s(x, y) is the feature covariance comparison value between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position. σ xy is the covariance of the corresponding pixel (eg, X(i, j)) in the positioning picture 1, and the corresponding pixel (eg, Y(i, j)) in the positioning picture 2. As an example, the covariance can be obtained by the following formula:
在获取了上述局部特征的对比值后,即可根据该对比值确定两个定位图片的相似度。After the comparison value of the above-mentioned local features is obtained, the similarity between the two positioning pictures can be determined according to the comparison value.
例如,可以根据如下公式(4)确定两个定位图片中,某一窗口位置的相似度。For example, the similarity of a certain window position in the two positioning pictures can be determined according to the following formula (4).
其中,S(x,y)为窗口处于某一位置时,与该位置对应的定位图片1和定位图片2之间的相似度。Wherein, S(x, y) is the similarity between the positioning picture 1 and the positioning picture 2 corresponding to the position when the window is at a certain position.
在根据上述公式(4)确定窗口处于定位图片中某一位置时的相似度后,可以根据统计的方法,确定整张定位图片之间的相似度。After determining the similarity when the window is at a certain position in the positioning picture according to the above formula (4), the similarity between the whole positioning pictures can be determined according to the statistical method.
例如,可以根据如下公式(5)确定定位图片1和定位图片2的相似度。For example, the similarity between the positioning picture 1 and the positioning picture 2 can be determined according to the following formula (5).
其中,MS(x,y)为定位图片1和定位图片2的相似度。M为每个定位图片中获取的局部特征对应位置的个数。j为M个位置中的任一个位置。μ
xj为窗口位于定位图片1中j对应位置时的特征均值。μ
yj为窗口位于定位图片2中j对应位置时的特征均值。σ
xj为窗口位于定位图片1中j对应位置时的特征方差。σ
yj为窗口位于定位图片2中j对应位置时的特征方差。σ
xjyj为协方差。
Among them, MS(x, y) is the similarity between the positioning picture 1 and the positioning picture 2. M is the number of corresponding positions of the local features obtained in each positioning image. j is any one of the M positions. μ xj is the feature mean when the window is located at the position corresponding to j in the positioning picture 1. μ yj is the feature mean value when the window is located at the position corresponding to j in the positioning picture 2. σ xj is the feature variance when the window is located at the position corresponding to j in the positioning picture 1. σ yj is the feature variance when the window is located at the position corresponding to j in the positioning picture 2. σ xjyj is the covariance.
在该示例中,可以根据MS(x,y)与第一阈值之间的大小关系,确定定位图片1和定位图片2是否处于相同的道路图层。例如,当MS(x,y)大于第一阈值时,则说明定位图片1和定位图片2的相似度较高,因此认为定位图片1和定位图片2为相同道路图层中的定位图片。如果当MS(x,y)小于第一阈值时,则说明定位图片1和定位图片2的相似度较低,因此认为定位图片1和定位图片2是覆盖相同水平区域但是位于不同道路图层的定位图片。作为一种示例,该第一阈值可以为0.5。In this example, it can be determined whether the positioning picture 1 and the positioning picture 2 are in the same road layer according to the magnitude relationship between MS(x, y) and the first threshold. For example, when MS(x, y) is greater than the first threshold, it means that the similarity between the positioning picture 1 and the positioning picture 2 is high, so the positioning picture 1 and the positioning picture 2 are considered to be the positioning pictures in the same road layer. If MS(x, y) is less than the first threshold, it means that the similarity between positioning picture 1 and positioning picture 2 is low, so it is considered that positioning picture 1 and positioning picture 2 cover the same horizontal area but are located in different road layers Position the picture. As an example, the first threshold may be 0.5.
需要说明的是,在该方法的说明中,定位图片的大小,窗口的大小以及滑动步长均为示例性说明,在其他一些实现方式中,定位图片的大小,窗口的大小以及滑动步长等参数均可根据实际需要灵活选取,本申请实施例对此不作限制。另外,本示例中,局部特征同时包括特征均值,特征方差以及特征协方差,参考这三个参数共同确定定位图片1和定位图片2之间的相似度。在另一些实现方式中,也可只参考特征均值,特征方差,以及特征协方差中的一个或任意两个,确定定位图片1和定位图片2的相似度。It should be noted that, in the description of this method, the size of the positioning image, the size of the window and the sliding step size are all exemplary descriptions. In some other implementations, the size of the positioning image, the size of the window and the sliding step size, etc. The parameters can be flexibly selected according to actual needs, which are not limited in this embodiment of the present application. In addition, in this example, the local features include feature mean, feature variance, and feature covariance at the same time, and the similarity between the positioning picture 1 and the positioning picture 2 is jointly determined with reference to these three parameters. In other implementations, only one or any two of the feature mean, feature variance, and feature covariance may be referred to to determine the similarity between the positioning picture 1 and the positioning picture 2.
可以理解的是,该局部纹理及全局参考法,通过对定位图片的局部特征进行详细的评估对比,以获取两个定位图片的相似度。因此,该相似度的度量结果更能体现定位图片中细节的差异,在定位图片对应的环境复杂度较低时,能够有效地根据局部特征判断两个定位图片的差异,因此在定位图片对应的环境较为简单时的相似度度量。It can be understood that the local texture and global reference method obtains the similarity between the two positioning pictures by performing detailed evaluation and comparison of the local features of the positioning pictures. Therefore, the similarity measurement result can better reflect the difference in the details of the positioning pictures. When the environment complexity corresponding to the positioning pictures is low, the difference between the two positioning pictures can be effectively judged according to the local features. A similarity measure when the environment is simpler.
2、地图特征指纹法。2. Map feature fingerprint method.
一般而言,定位图片中的像素特征个数可能较多。在采用该地图特征指纹法评估两个定位图片的相似度时,可以分别将两个定位图片缩小到较小的尺寸,并根据缩小后的图片中的每个像素所表征的相对高度,计算获取对应定位图片的高度均值。根据该高度均值对定位图片进行归一化处理,由此得到两个定位图片各自的高度特征。通过对比两个定位图片的高度特征,确定两个定位图片的相似度。Generally speaking, the number of pixel features in the positioning image may be large. When using the map feature fingerprinting method to evaluate the similarity of two positioning pictures, the two positioning pictures can be respectively reduced to a smaller size, and the relative height represented by each pixel in the reduced picture can be calculated and obtained. The average height of the corresponding positioning image. The positioning picture is normalized according to the height mean value, thereby obtaining the respective height features of the two positioning pictures. By comparing the height features of the two positioning pictures, the similarity of the two positioning pictures is determined.
示例性的,首先对两个定位图片进行处理。对于定位图片1和定位图片2中的每个图片,可以通过隔行/隔列删除像素的方法,缩小定位图片的尺寸。例如,将定位图片缩小到如图9中的(a)所示的3*3的像素尺寸。为了便于说明,以a
ij标识缩小后的图片中第i行第j列的像素,i和j均为小于或等于3的正整数。在获取如图9中的(a)所示的缩小后的去除了定位图片中的细节信息,只保留基本结构以及明暗信息的具有较小尺寸的图片后,可以根据每个像素的灰度,确定对应像素的相对高度。如a
11的相对高度为h
11,a
12的相对高度为h
12,a
ij的相对高度为h
ij。根据每个像素的相对高 度,即可获取该缩小后的图片的每个像素的高度均值Δh。根据该Δh,即可对对应的定位图片进行归一化处理。例如,可以通过以下方法进行归一化处理:对比h
ij与Δh的大小关系,如果h
ij大于Δh,则将对应的a
ij标识为1。反之,如果h
ij小于Δh,则将对应的a
ij标识为0。这样就可以获取一个3*3的矩阵。其中,每个元素均为0或1。例如,可以获取如图9中的(b)所示的矩阵。将该矩阵中的元素顺序排列即可获取一个9比特大小的二进制数。以矩阵为如图9中的(b)所示为例,顺序排列后对应的二进制数可以为101001111。由此,即可将该二进制数称为对应定位图片的特征指纹。
Exemplarily, the two positioning pictures are first processed. For each picture in the positioning picture 1 and the positioning picture 2, the size of the positioning picture can be reduced by deleting pixels in every row/column. For example, the positioning picture is reduced to a pixel size of 3*3 as shown in (a) of FIG. 9 . For the convenience of description, the pixel in the i-th row and the j-th column in the reduced picture is identified by a ij , where i and j are both positive integers less than or equal to 3. After obtaining a reduced image with a smaller size as shown in (a) in FIG. 9 , the detail information in the positioning image is removed, and only the basic structure and light and shade information are retained, according to the grayscale of each pixel, Determines the relative height of the corresponding pixel. For example, the relative height of a 11 is h 11 , the relative height of a 12 is h 12 , and the relative height of a ij is h ij . According to the relative height of each pixel, the average height Δh of each pixel of the reduced picture can be obtained. According to the Δh, the corresponding positioning picture can be normalized. For example, normalization processing can be performed by the following method: comparing the magnitude relationship between h ij and Δh, if h ij is greater than Δh, the corresponding a ij is marked as 1. Conversely, if h ij is less than Δh, the corresponding a ij is marked as 0. In this way, a 3*3 matrix can be obtained. where each element is either 0 or 1. For example, a matrix as shown in (b) in FIG. 9 can be acquired. Arrange the elements in the matrix in order to obtain a 9-bit binary number. Taking the matrix as shown in (b) in FIG. 9 as an example, the corresponding binary number after the sequential arrangement may be 101001111. Thus, the binary number can be called the feature fingerprint of the corresponding positioning picture.
应当理解的是,基于上述方法,可以获取定位图片1和定位图片2分别对应的特征指纹。通过对比两个图片的特征指纹,即可确定两个图片的相似度。It should be understood that, based on the above method, the feature fingerprints respectively corresponding to the positioning picture 1 and the positioning picture 2 can be obtained. By comparing the feature fingerprints of the two images, the similarity of the two images can be determined.
例如,以定位图片1的特征指纹为101001111,定位图片2的特征指纹为101010010为例。可以通过计算两个特征指纹的汉明距离(hamming distance),确定两个定位图片的相似度。结合图10,定位图片1的特征指纹和定位图片2的特征指纹中,存在如图中的虚线框中所示的3位不同,则两个定位图片的特征指纹的汉明距离为3。可以理解的是,汉明距离越大,则两个定位图片的相似度越低。汉明距离越小,则两个定位图片的相似度越高。本申请中,采用缩小到8×8特征地图,可以对比该汉明距离与第二阈值之间的大小关系,确定两个定位图片是否处于相同的道路图层。例如,当汉明距离大于第二阈值时,则认为两个定位图片不在同一个道路图层。又如当汉明距离小于第二阈值时,则认为两个定位图片属于同一个道路图层。作为一种示例,该第二阈值可以为8。在本申请中,可以将根据特征指纹确定的相似度称为指纹相似度(map fingerprint similarity,mfs)。For example, take the feature fingerprint of positioning picture 1 as 101001111 and the feature fingerprint of positioning picture 2 as 101010010 as an example. The similarity of the two positioning images can be determined by calculating the Hamming distance of the two feature fingerprints. Referring to FIG. 10 , the feature fingerprint of the positioning picture 1 and the feature fingerprint of the positioning picture 2 are different by 3 bits as shown in the dotted box in the figure, and the Hamming distance of the feature fingerprints of the two positioning pictures is 3. It can be understood that the larger the Hamming distance, the lower the similarity between the two positioning images. The smaller the Hamming distance, the higher the similarity between the two positioning images. In the present application, by using the feature map reduced to 8×8, the size relationship between the Hamming distance and the second threshold can be compared to determine whether the two positioning pictures are in the same road layer. For example, when the Hamming distance is greater than the second threshold, it is considered that the two positioning images are not in the same road layer. For another example, when the Hamming distance is less than the second threshold, it is considered that the two positioning images belong to the same road layer. As an example, the second threshold may be 8. In this application, the similarity determined according to the feature fingerprint may be referred to as the fingerprint similarity (map fingerprint similarity, mfs).
可以理解的是,该地图特征指纹法,通过筛除定位图片中的细节部分,保留基本信息,并从全局对定位图片进行相似度度量,能够更加体现定位图片的全局差异。因此更加适用于定位图片对应的环境较为复杂时的相似度度量。It can be understood that the map feature fingerprint method can better reflect the global differences of the positioning pictures by filtering out the details in the positioning pictures, retaining the basic information, and measuring the similarity of the positioning pictures from the global perspective. Therefore, it is more suitable for similarity measurement when the environment corresponding to the positioning picture is relatively complex.
需要说明的是,上述示例中提供的两种相似度度量方法(如局部纹理及全局参考法,地图特征指纹法),可以根据场景的不同灵活选取其一使用,也可同时采用这两种方法进行自适应的相似度度量。It should be noted that the two similarity measurement methods provided in the above example (such as local texture and global reference method, map feature fingerprint method) can be flexibly selected and used according to different scenes, or both methods can be used at the same time. Perform an adaptive similarity measure.
当同时采用这两种方法进行相似度度量时,可以引入用于表征环境复杂度的参数(如用α标识),来调整两种方法分别确定的相似度在结果中的权重,以便根据环境复杂度的变化,自适应地进行相似度度量。When these two methods are used to measure the similarity at the same time, the parameters used to characterize the complexity of the environment (such as marked with α) can be introduced to adjust the weight of the similarity determined by the two methods respectively in the result, so that according to the complexity of the environment The degree of similarity is measured adaptively.
示例性的,可以通过如下公式(6)最终的相似度。Exemplarily, the final similarity can be obtained by the following formula (6).
MapS=f(MS,mfs,α)……公式(6)。MapS=f(MS, mfs, α) … Equation (6).
其中,MapS为定位图片1和定位图片2的相似度。MS为根据局部纹理及全局参考法确定的相似度。mfs为根据地图特征指纹法确定的相似度。α为环境复杂度。Among them, MapS is the similarity between the positioning picture 1 and the positioning picture 2. MS is the similarity determined according to the local texture and global reference method. mfs is the similarity determined according to the map feature fingerprinting method. α is the environment complexity.
在不同的场景下,该公式(6)可以是灵活设置的。例如,在一些实现方式中,该公式(6)可以变换为如下公式(6-1)。In different scenarios, the formula (6) can be set flexibly. For example, in some implementations, the formula (6) can be transformed into the following formula (6-1).
MapS=α×MS+(1-α)×mfs……公式(6-1)。MapS=α×MS+(1−α)×mfs ...... Formula (6-1).
其中,α可以根据环境复杂度的变化,灵活设置为小于或等于1且大于或等于0的数。例如,当定位图片对应的环境较为复杂时,可以全局相似度为主进行相似度度 量,因此可以将α设置在(0.5,1]的范围内,可以提升MS的权重。又如,当定位图片对应的环境较为简单时,可以局部细节的相似度为主进行相似度度量,因此可以将α设置在[0,0.5)的范围内,可以提升mfs的权重。由此即可实现根据环境复杂度的不同,自适应地调整定位图片的相似度度量的目的。Wherein, α can be flexibly set to a number less than or equal to 1 and greater than or equal to 0 according to changes in the complexity of the environment. For example, when the environment corresponding to the positioning picture is relatively complex, the global similarity can be used as the main similarity measure, so α can be set in the range of (0.5, 1], which can improve the weight of MS. For another example, when positioning the picture When the corresponding environment is relatively simple, the similarity of local details can be mainly used for similarity measurement. Therefore, α can be set in the range of [0, 0.5), which can increase the weight of mfs. In this way, the purpose of adaptively adjusting the similarity measure of the positioning pictures according to the different complexity of the environment can be achieved.
在确定MapS后,可以将该MapS与第三阈值进行对比,确定定位图片1和定位图片2是否处于相同的道路图层。例如,如果MapS大于第三阈值,则认为定位图片1和定位图片2之间的相似度较高,属于相同的道路图层。如果MapS小于第三阈值,则认为定位图片1和定位图片2之间的相似度较低,属于不同的道路图层。After the MapS is determined, the MapS can be compared with the third threshold to determine whether the positioning picture 1 and the positioning picture 2 are in the same road layer. For example, if MapS is greater than the third threshold, it is considered that the similarity between the positioning picture 1 and the positioning picture 2 is high and belong to the same road layer. If MapS is less than the third threshold, it is considered that the similarity between the positioning picture 1 and the positioning picture 2 is low, and they belong to different road layers.
需要说明的是,在上述示例中,可以将覆盖相同水平区域的定位图片作为相似度度量的对象,以便确定不同的定位图片是否是车辆在不同时刻相同道路上采集获取的。在另一些示例中,还可将多个定位图片中,任意两个定位图片作为相似度度量的对象,以便通过该方法,准确地确定多个定位图片中的任意两个定位图片是否处于同一个道路图层。It should be noted that, in the above example, the positioning pictures covering the same horizontal area can be used as the object of similarity measurement, so as to determine whether different positioning pictures are collected by vehicles on the same road at different times. In other examples, any two positioning pictures in the multiple positioning pictures can also be used as the objects of similarity measurement, so that through this method, it can be accurately determined whether any two positioning pictures in the multiple positioning pictures are in the same one road layer.
为了使得本领域技术人员能够更加清楚地理解上述示例中提供的相似度度量方案及其有益效果,以下结合实际试验,对同时采用局部纹理及全局参考法以及地图特征指纹法进行相似度度量的过程和结果进行示例性说明。In order to enable those skilled in the art to more clearly understand the similarity measurement scheme provided in the above example and its beneficial effects, the following combined with actual experiments, the process of using the local texture and global reference method and the map feature fingerprint method to simultaneously measure the similarity and results are exemplified.
在实验1中,以环境特征较为复杂,定位图片1和定位图片2为根据相同道路图层中,不同时间采集获取的激光点云数据获取为例。结合图11,定位图片1和定位图片2如图中所示。根据上述示例中的方法,计算获取两个定位图片对应的特征指纹的汉明距离为3,基于局部纹理及全局参考法获取的相似度MS为0.6192。可以根据环境复杂度,调整两个相似度的权重,最终得到两个的定位图片的相似度为0.785823。In experiment 1, the environment features are more complex, and the positioning picture 1 and positioning picture 2 are based on the laser point cloud data obtained in the same road layer and collected at different times as an example. With reference to FIG. 11 , the positioning picture 1 and the positioning picture 2 are shown in the figure. According to the method in the above example, the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 3, and the similarity MS obtained based on the local texture and the global reference method is 0.6192. The weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.785823.
在实验2中,以环境特征较为简单,定位图片1和定位图片2为根据相同道路图层中,不同时间采集获取的激光点云数据获取为例。结合图12,定位图片1和定位图片2如图中所示。根据上述示例中的方法,计算获取两个定位图片对应的特征指纹的汉明距离为1,基于局部纹理及全局参考法获取的相似度MS为0.885771。可以根据环境复杂度,调整两个相似度的权重,最终得到两个的定位图片的相似度为0.927816。In experiment 2, the environmental characteristics are relatively simple, and the positioning picture 1 and the positioning picture 2 are obtained according to the laser point cloud data collected at different times in the same road layer as an example. With reference to FIG. 12 , the positioning picture 1 and the positioning picture 2 are shown in the figure. According to the method in the above example, the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 1, and the similarity MS obtained based on the local texture and the global reference method is 0.885771. The weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.927816.
在实验3中,以环境特征较为复杂,定位图片1和定位图片2为根据不同道路图层中采集获取的激光点云数据获取为例。结合图13,定位图片1和定位图片2如图中所示。根据上述示例中的方法,计算获取两个定位图片对应的特征指纹的汉明距离为13,基于局部纹理及全局参考法获取的相似度MS为0.051161。可以根据环境复杂度,调整两个相似度的权重,最终得到两个的定位图片的相似度为0.085639。In experiment 3, the environment features are more complex, and the positioning picture 1 and positioning picture 2 are obtained from laser point cloud data collected from different road layers as an example. With reference to FIG. 13 , the positioning picture 1 and the positioning picture 2 are shown in the figure. According to the method in the above example, the Hamming distance calculated to obtain the feature fingerprints corresponding to the two positioning pictures is 13, and the similarity MS obtained based on the local texture and the global reference method is 0.051161. The weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.085639.
在实验4中,以环境特征较为简单,定位图片1和定位图片2为根据不同道路图层中采集获取的激光点云数据获取为例。结合图14,定位图片1和定位图片2如图中所示。根据上述示例中的方法,计算获取两个定位图片对应的特征指纹的汉明距离为0,基于局部纹理及全局参考法获取的相似度MS为0.133746。可以根据环境复杂度,调整两个相似度的权重,最终得到两个的定位图片的相似度为0.165378。In experiment 4, the environment features are relatively simple, and the positioning picture 1 and the positioning picture 2 are obtained from the laser point cloud data collected from different road layers as an example. With reference to FIG. 14 , the positioning picture 1 and the positioning picture 2 are shown in the figure. According to the method in the above example, the Hamming distance of the characteristic fingerprints corresponding to the two positioning pictures is calculated to be 0, and the similarity MS obtained based on the local texture and the global reference method is 0.133746. The weights of the two similarities can be adjusted according to the complexity of the environment, and the similarity of the two positioning images is finally obtained as 0.165378.
对比上述实验1和实验2两个实验结果,可以看到,当环境特征较为复杂时,汉明距离较大,基于局部纹理及全局参考法获取的相似度较小。而两个实验的前提限定了定位图片1和定位图片2为同一道路图层中的定位图片。因此,当环境特征较为复 杂时,基于局部纹理及全局参考法获取的相似度更能够准确地体现两个定位图片的相似度情况。也就是说,在环境特征较为复杂时,如果采用单一方法进行相似度度量,可以优先考虑使用基于局部纹理及全局参考法获取的相似度来判断两个定位图片是否处于同一道路图层。如果同时采用两种方法进行相似度度量,可以通过合理调整两种方法获取的相似度的权重,如增加基于局部纹理及全局参考法获取的相似度的权重,以便能够获取更加接近实际的相似度度量结果。Comparing the above two experimental results of Experiment 1 and Experiment 2, it can be seen that when the environmental features are more complex, the Hamming distance is larger, and the similarity obtained based on the local texture and global reference method is smaller. The premise of the two experiments limits the positioning picture 1 and the positioning picture 2 to be the positioning pictures in the same road layer. Therefore, when the environmental features are more complex, the similarity obtained based on the local texture and the global reference method can more accurately reflect the similarity of the two positioning pictures. That is to say, when the environmental features are complex, if a single method is used to measure the similarity, the similarity obtained based on the local texture and the global reference method can be prioritized to judge whether the two positioning pictures are in the same road layer. If two methods are used for similarity measurement at the same time, the weight of the similarity obtained by the two methods can be adjusted reasonably, such as increasing the weight of the similarity obtained based on the local texture and the global reference method, so as to obtain the similarity closer to the actual Measure the results.
对应的,当环境特征较为简单时,汉明距离较小,基于局部纹理及全局参考法获取的相似度较大。而两个实验的前提限定了定位图片1和定位图片2为同一道路图层中的定位图片。因此,当环境特征较为简单时,基于地图特征指纹法获取的相似度更能够准确地体现两个定位图片的相似度情况。也就是说,在环境特征较为简单时,如果采用单一方法进行相似度度量,可以优先考虑使用地图特征指纹法获取的相似度来判断两个定位图片是否处于同一道路图层。如果同时采用两种方法进行相似度度量,可以通过合理调整两种方法获取的相似度的权重,如增加地图特征指纹法获取的相似度的权重,以便能够获取更加接近实际的相似度度量结果。Correspondingly, when the environmental features are relatively simple, the Hamming distance is small, and the similarity obtained based on the local texture and global reference method is large. The premise of the two experiments limits the positioning picture 1 and the positioning picture 2 to be the positioning pictures in the same road layer. Therefore, when the environmental features are relatively simple, the similarity obtained based on the map feature fingerprinting method can more accurately reflect the similarity of the two positioning pictures. That is to say, when the environmental features are relatively simple, if a single method is used to measure the similarity, the similarity obtained by the map feature fingerprinting method can be prioritized to judge whether the two positioning pictures are in the same road layer. If two methods are used to measure the similarity at the same time, the weight of the similarity obtained by the two methods can be adjusted reasonably, such as increasing the weight of the similarity obtained by the map feature fingerprinting method, so as to obtain the similarity measurement result that is closer to the actual.
类似的,通过对比实验3和实验4也能得出相同的结论。同时,通过对比实验1和3,以及实验2和4,能够明确地看到,通过本申请实施例提供的两种方法,都能够准确地分辨出两个定位图片是否处于同一个道路图层中。Similarly, the same conclusion can be drawn by comparing Experiment 3 and Experiment 4. At the same time, by comparing Experiments 1 and 3, and Experiments 2 and 4, it can be clearly seen that whether the two positioning pictures are in the same road layer can be accurately distinguished through the two methods provided in the embodiments of the present application .
另外,在确定两个定位图片属于同一道路图层后,可以将这两个定位图片进行融合处理,以获取具有更加准确的细节信息的定位图片。例如,可以将实验1中的两个图片进行融合。其融合结果如图15中的(a)所示。对比如图15中的(a)所示的定位图片和实验1中的两个定位图片,可以看到融合后的定位图片具有更加准确的细节信息,因此,该融合处理有助于弥补单次采集的地图缺失。类似的,将实验2中的两个图片进行融合后的结果如图15中的(b)所示,与融合之前的图片相比,也具有类似的特征。In addition, after it is determined that the two positioning pictures belong to the same road layer, the two positioning pictures may be fused to obtain a positioning picture with more accurate detailed information. For example, the two pictures in Experiment 1 can be fused. The fusion result is shown in (a) of FIG. 15 . Comparing the positioning picture shown in (a) in Figure 15 with the two positioning pictures in Experiment 1, it can be seen that the fused positioning picture has more accurate detailed information. Therefore, the fusion process helps to compensate for the single-shot The collected map is missing. Similarly, the result of merging the two pictures in Experiment 2 is shown in (b) in Figure 15, which also has similar characteristics compared with the pictures before fusion.
根据上述说明,在确定了定位图片的层间关系(如属于同一个道路图层,或属于不同道路图层)后,可以将处于相同道路图层的定位图片进行融合,以获取对应的道路图层的定位地图。在一些实现方式中,可以对不同的道路图层打上不同的地图标签(如map id)以便在车辆进行自动驾驶过程中,可以根据当前自身的位置信息(如通过GPS定位***确定的当前在全局坐标系下的高度信息),选取对应map id的定位地图,以获取准确的定位信息。According to the above description, after determining the inter-layer relationship of the positioning pictures (for example, belonging to the same road layer or belonging to different road layers), the positioning pictures in the same road layer can be fused to obtain the corresponding road map The location map of the layer. In some implementations, different map labels (such as map id) can be added to different road layers, so that during the automatic driving process of the vehicle, it can be based on the current position information (such as the current global location determined by the GPS positioning system). height information in the coordinate system), select the positioning map corresponding to the map id to obtain accurate positioning information.
另外,需要说明的是,在本申请实施例的说明中,当两个具有相同水平覆盖区域的定位图片处于相同道路图层(如确定这两个定位图片的相似度较高)时,则可以对这两个定位图片进行融合处理。在不同的实现方式中,该融合处理可以包括多种不同的具体实现。示例性的,在一些实现方式中,确定两个具有相同水平覆盖其余的定位图片处于相同的道路图层时,可以针对这两个定位图片中,每个对应位置的两个像素进行融合,以获取融合后的一个像素。以此类推可以获取融合处理后的多个位置的像素,这些像素即可构成融合处理后的定位图片。其中,该针对两个像素的融合,可以是对两个像素的像素值(如灰度值)取均值后作为融合之后的像素的像素值。也可以是对两个像素的像素值取最大值作为融合之后的像素的像素值。当然,也可是对两个 像素的像素值取最小值作为融合之后的像素的像素值。本申请实施例对于像素的融合处理机制不做限制。In addition, it should be noted that, in the description of the embodiments of the present application, when two positioning pictures with the same horizontal coverage area are in the same road layer (for example, it is determined that the similarity between the two positioning pictures is high), then the The two positioning images are fused. In different implementations, the fusion process may include various specific implementations. Exemplarily, in some implementations, when it is determined that two positioning pictures with the same horizontal coverage and the rest are in the same road layer, two pixels at each corresponding position in the two positioning pictures may be fused to Get a fused pixel. By analogy, the pixels of multiple positions after fusion processing can be obtained, and these pixels can constitute the positioning image after fusion processing. Wherein, for the fusion of two pixels, the pixel values of the two pixels (eg, gray values) can be averaged as the pixel value of the pixel after fusion. It is also possible to take the maximum value of the pixel values of two pixels as the pixel value of the pixel after fusion. Of course, the minimum value of the pixel values of the two pixels can also be taken as the pixel value of the pixel after fusion. The embodiments of the present application do not limit the pixel fusion processing mechanism.
在上述处于相同道路图层的两个定位图片进行融合操作之外,在本申请实施例中,确定第一定位图片的集合以及第二定位图片的集合之后,可以分别对第一定位图片的集合中的第一定位图片进行融合处理,以获取与第一道路图层对应的定位地图。示例性的,对于具有不同水平覆盖区域的定位地图的融合处理,可以根据定位图片中,处于边缘的像素点对应的水平坐标进行的。例如,定位图片A的左下角的像素在全局坐标系下的水平面坐标为(Xa,Ya),那么可以将右下角的像素坐标为(Xa,Ya)的定位图片B与该定位图片A进行融合,该融合的过程可以为将定位图片A放置在定位图片B相邻的右边以融合获取一个新的同时包括定位图片A和定位图片B的定位图片。以此类推,将所有第一定位图片的集合中的定位图片执行上述融合操作,即可获取一张与第一道路对应的完整的定位地图。需要说明的是,在实际实现过程中,也可以在确定第一定位图片的集合和第二定位图片的集合之后,不对进行融合,而是分别将该定位图片的集合存储在服务器,或者下发给需要使用的设备(如车辆)。在车辆需要使用对应的道路图层的地图时,结合自身位置,对对应道路图层中的与车辆所在位置相邻的定位图片进行融合,后去该道路图层对应的局部定位地图。以便为车辆当前的行驶提供准确的定位信息。In addition to the above-mentioned fusion operation of the two positioning pictures in the same road layer, in this embodiment of the present application, after the set of the first positioning picture and the set of the second positioning picture are determined, the set of the first positioning picture can be respectively Fusion processing is performed on the first positioning picture in the first road layer to obtain a positioning map corresponding to the first road layer. Exemplarily, the fusion processing of positioning maps with different horizontal coverage areas may be performed according to the horizontal coordinates corresponding to the pixels at the edge in the positioning picture. For example, the horizontal plane coordinates of the pixel in the lower left corner of the positioning picture A in the global coordinate system are (Xa, Ya), then the positioning picture B whose pixel coordinates in the lower right corner are (Xa, Ya) can be fused with the positioning picture A. , the fusion process may be to place the positioning picture A on the adjacent right side of the positioning picture B to obtain a new positioning picture that includes the positioning picture A and the positioning picture B at the same time. By analogy, by performing the above fusion operation on the positioning pictures in the set of all the first positioning pictures, a complete positioning map corresponding to the first road can be obtained. It should be noted that, in the actual implementation process, after the set of the first positioning picture and the set of the second positioning picture are determined, the set of positioning pictures is not fused, but the set of positioning pictures is stored in the server, or sent to the server. For equipment that needs to be used (eg vehicles). When the vehicle needs to use the map of the corresponding road layer, it combines its own position, fuses the positioning pictures in the corresponding road layer that are adjacent to the vehicle's location, and then goes to the local positioning map corresponding to the road layer. In order to provide accurate positioning information for the current driving of the vehicle.
上述主要从设备(如用于绘制定位地图的设备)的角度对本申请实施例提供的方案进行了介绍。为了实现上述功能,其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。The foregoing mainly introduces the solutions provided by the embodiments of the present application from the perspective of a device (eg, a device for drawing a positioning map). In order to realize the above-mentioned functions, it includes corresponding hardware structures and/or software modules for executing each function. Those skilled in the art should easily realize that the present application can be implemented in hardware or a combination of hardware and computer software with the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein. Whether a function is performed by hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.
本申请实施例可以根据上述方法示例对其中涉及的设备进行功能模块的划分,例如,可以对应各个功能划分各个功能模块,也可以将两个或两个以上的功能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现,也可以采用软件功能模块的形式实现。需要说明的是,本申请实施例中对模块的划分是示意性的,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。In this embodiment of the present application, the devices involved may be divided into functional modules according to the above method examples. For example, each functional module may be divided into each function, or two or more functions may be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. It should be noted that, the division of modules in the embodiments of the present application is schematic, and is only a logical function division, and there may be other division manners in actual implementation.
请参考图16,为本申请实施例提供的一种地图绘制装置1600的组成示意图。该地图绘制装置1600可以是终端,也可以是终端内部的芯片,并且可以实现如图3以及上述各可选实施例。Please refer to FIG. 16 , which is a schematic diagram of the composition of a map drawing apparatus 1600 according to an embodiment of the present application. The map drawing apparatus 1600 may be a terminal, or a chip inside the terminal, and may implement the optional embodiments as shown in FIG. 3 and the above.
如图16所示,该地图绘制装置可包括:获取单元1601,融合单元1602。其中,获取单元1601可以用于执行如图3所示方法中的S301-S303中任一步骤以及其中任一可选的实施例。融合单元1602可以用于执行如图3所示的S304中任一步骤以及其中任一可选的实施例。As shown in FIG. 16 , the map drawing apparatus may include: an acquisition unit 1601 and a fusion unit 1602 . Wherein, the obtaining unit 1601 may be configured to perform any one of the steps S301 to S303 in the method shown in FIG. 3 and any one of the optional embodiments. The fusion unit 1602 may be configured to perform any of the steps in S304 shown in FIG. 3 and any of the optional embodiments.
示例性的,该获取单元1601,用于获取第一待绘制区域的激光点云数据,第一待绘制区域包括处于不同平面的第一道路和第二道路。Exemplarily, the acquiring unit 1601 is configured to acquire laser point cloud data of a first area to be drawn, where the first area to be drawn includes a first road and a second road on different planes.
该获取单元1601,还用于根据激光点云数据,获取第一定位图片的集合和第二定 位图片的集合。其中,任意两个第一定位图片的参考高度相差不超过第一阈值。任意两个第二定位图片的参考高度相差不超过第一阈值。第一定位图片和第二定位图片的参考高度差大于第二阈值。定位图片的参考高度是采集定位图片对应的激光点云数据时,对应设备所在路面的高度。第一阈值和第二阈值均为正数。The obtaining unit 1601 is further configured to obtain a set of first positioning pictures and a set of second positioning pictures according to the laser point cloud data. Wherein, the difference between the reference heights of any two first positioning pictures does not exceed the first threshold. The difference between the reference heights of any two second positioning pictures does not exceed the first threshold. The reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold. The reference height of the positioning picture is the height of the road corresponding to the device when the laser point cloud data corresponding to the positioning picture is collected. Both the first threshold and the second threshold are positive numbers.
融合单元1602,用于将第一定位图片集合中的定位图片进行融合,获得第一道路对应的地图,将第二定位图片集合中的定位图片进行融合,获得第二道路对应的地图。The fusion unit 1602 is configured to fuse the positioning pictures in the first positioning picture set to obtain a map corresponding to the first road, and fuse the positioning pictures in the second positioning picture set to obtain a map corresponding to the second road.
在一种可能的设计中,该激光点云数据包括:第一位置在全局坐标系下的三维坐标信息,其中,第一位置为采集激光点云数据的位置。该获取单元1601,还用于根据第一位置的三维坐标信息,获取第一位置相对于采集激光点云数据时所行驶道路的相对高度信息。In a possible design, the laser point cloud data includes: three-dimensional coordinate information of the first position in the global coordinate system, where the first position is the position where the laser point cloud data is collected. The obtaining unit 1601 is further configured to obtain, according to the three-dimensional coordinate information of the first position, the relative height information of the first position relative to the road on which the laser point cloud data is collected.
在一种可能的设计中,该激光点云数据还包括:用于指示第一位置是否为车道的第一标识。In a possible design, the laser point cloud data further includes: a first sign used to indicate whether the first position is a lane.
在一种可能的设计中,该获取单元1601,用于根据激光点云数据,获取多个定位图片,每个定位图片对应的激光点云数据的采集时间在预设范围之内,每个定位图片中包括的像素的像素值由像素对应位置的相对高度信息和位置对应的激光点云数据的第一标识确定。该融合单元1602,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,获取第一定位图片集合和第二定位图片集合。In a possible design, the acquiring unit 1601 is configured to acquire a plurality of positioning pictures according to the laser point cloud data, and the collection time of the laser point cloud data corresponding to each positioning picture is within a preset range, and each positioning picture The pixel value of the pixel included in the picture is determined by the relative height information of the corresponding position of the pixel and the first identifier of the laser point cloud data corresponding to the position. The fusion unit 1602 is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set and a second positioning picture set.
在一种可能的设计中,该装置还包括:确定单元1603,该确定单元1603用于确定第一定位图片和第二定位图片的相似度。该确定单元1603,还用于根据相似度,确定第一定位图片和第二定位图片是否处于相同道路图层,相似度用于指示第一定位图片和第二定位图片的相似程度。其中,第一定位图片和第二定位图片为多个定位图片中,任意两个具有相同水平覆盖区域的定位图片。融合单元1602,用于将多个定位图片中,处于相同道路图层的定位图片进行融合,以获取处于第一道路对应的第一定位图片集合,以及第二道路对应的第二定位图片集合。In a possible design, the apparatus further includes: a determining unit 1603, where the determining unit 1603 is configured to determine the similarity between the first positioning picture and the second positioning picture. The determining unit 1603 is further configured to determine whether the first positioning picture and the second positioning picture are in the same road layer according to the similarity, where the similarity is used to indicate the degree of similarity between the first positioning picture and the second positioning picture. Wherein, the first positioning picture and the second positioning picture are any two positioning pictures having the same horizontal coverage area among the multiple positioning pictures. The fusion unit 1602 is configured to fuse the positioning pictures in the same road layer among the multiple positioning pictures to obtain a first positioning picture set corresponding to the first road and a second positioning picture set corresponding to the second road.
在一种可能的设计中,该装置还包括:确定单元1603,该确定单元1603用于根据第一定位图片和第二定位图片的局部特征,确定第一定位图片和第二定位图片的第一相似度。局部特征包括以下中的一项或多项:定位图片中像素的灰度平均值,像素的灰度方差,像素的灰度协方差。该确定单元1603,还用于当第一相似度大于第一阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the apparatus further includes: a determining unit 1603, the determining unit 1603 is configured to determine the first positioning picture and the second positioning picture according to the local features of the first positioning picture and the second positioning picture. similarity. The local features include one or more of the following: the grayscale mean of the pixels in the location image, the grayscale variance of the pixels, and the grayscale covariance of the pixels. The determining unit 1603 is further configured to determine that the first positioning picture and the second positioning picture are in the same road layer when the first similarity is greater than the first threshold.
在一种可能的设计中,确定单元1603,还用于当第一相似度小于第一阈值时,确定第一定位图片和第二定位图片处于不同道路图层。In a possible design, the determining unit 1603 is further configured to determine that the first positioning picture and the second positioning picture are in different road layers when the first similarity is less than the first threshold.
在一种可能的设计中,确定单元1603,还用于根据第一定位图片和第二定位图片中对应像素的相对高度信息,确定第一定位图片和第二定位图片的第二相似度。当第二相似度小于第二阈值时,确定第一定位图片和第二定位图片处于相同道路图层。In a possible design, the determining unit 1603 is further configured to determine the second similarity between the first positioning picture and the second positioning picture according to the relative height information of the corresponding pixels in the first positioning picture and the second positioning picture. When the second similarity is less than the second threshold, it is determined that the first positioning picture and the second positioning picture are in the same road layer.
在一种可能的设计中,该确定单元1603,还用于当第二相似度大于第二阈值时,第一定位图片和第二定位图片处于不同道路图层。In a possible design, the determining unit 1603 is further configured to, when the second similarity is greater than the second threshold, the first positioning picture and the second positioning picture are in different road layers.
在一种可能的设计中,获取单元1601,具体用于针对第一定位图片和第二定位图片分别执行以下操作,以获取第一定位图片对应的特征指纹,和第二定位图片对应的特征指纹:删除定位图片中预设行和/或预设列的像素,以获取缩小的定位图片,根据 缩小的定位图片中,各个像素的相对高度的均值,对缩小的定位图片进行归一化处理,根据归一化处理后的缩小的定位图片的各个像素值,确定定位图片对应的特征指纹。根据第一定位图片的特征指纹和第二定位图片的特征指纹,确定第一定位图片和第二定位图片的第二相似度,第二相似度为第一定位图片的特征指纹和第二定位图片的特征指纹的汉明距离。In a possible design, the obtaining unit 1601 is specifically configured to perform the following operations on the first positioning picture and the second positioning picture respectively, so as to obtain the feature fingerprint corresponding to the first positioning picture and the feature fingerprint corresponding to the second positioning picture : Delete the pixels of the preset row and/or preset column in the positioning picture to obtain a reduced positioning picture, and normalize the reduced positioning picture according to the mean value of the relative heights of each pixel in the reduced positioning picture, According to each pixel value of the normalized reduced positioning picture, the feature fingerprint corresponding to the positioning picture is determined. According to the feature fingerprint of the first positioning picture and the feature fingerprint of the second positioning picture, the second similarity between the first positioning picture and the second positioning picture is determined, and the second similarity is the feature fingerprint of the first positioning picture and the second positioning picture. The Hamming distance of the characteristic fingerprint.
应理解的是,本申请实施例中的地图绘制装置可以由软件实现,例如,具有上述功能的计算机程序或指令来实现,相应计算机程序或指令可以存储在终端内部的存储器中,通过处理器读取该存储器内部的相应计算机程序或指令来实现上述功能。或者,本申请实施例中的地图绘制装置还可以由硬件来实现。示例性的,获取单元1601和/或融合单元1602和/或确定单元1603可以通过处理器(如NPU、GPU、***芯片中的处理器)实现其对应的功能。或者,本申请实施例中的地图绘制装置还可以由处理器和软件模块的结合实现。It should be understood that the map drawing device in the embodiment of the present application may be implemented by software, for example, a computer program or instruction having the above-mentioned functions, and the corresponding computer program or instruction may be stored in the internal memory of the terminal, and read by the processor. The above-mentioned functions are realized by fetching the corresponding computer programs or instructions inside the memory. Alternatively, the map drawing apparatus in this embodiment of the present application may also be implemented by hardware. Exemplarily, the acquiring unit 1601 and/or the fusion unit 1602 and/or the determining unit 1603 may implement their corresponding functions through a processor (eg, an NPU, a GPU, or a processor in a system chip). Alternatively, the map drawing apparatus in this embodiment of the present application may also be implemented by a combination of a processor and a software module.
具体地,获取单元1601可以为处理器的接口电路。作为一种示例,接口电路可以将获取的多个激光点云数据传输给处理器。处理器可以用于对来自接口电路的激光点云数据进行预处理(如执行如图3所示的S302及该步骤中任一项可能的操作),处理器还可以用于根据预处理后的激光点云数据获取多个定位图片(如执行如图3所示的S303及该步骤中的任一项可能的操作),处理器还可以用于根据该多个定位图片获取不同道路图层对应的定位地图(如执行如图3所示的S304及该步骤中的任一项可能的操作)。应当理解的是,该处理器还可用于执行上述实施例中的其他操作,以便实现本申请实施例提供的任一种地图绘制方法。Specifically, the obtaining unit 1601 may be an interface circuit of the processor. As an example, the interface circuit may transmit the acquired multiple laser point cloud data to the processor. The processor can be used to preprocess the laser point cloud data from the interface circuit (such as performing S302 shown in FIG. 3 and any possible operation in this step), and the processor can also be used to The laser point cloud data obtains multiple positioning pictures (such as performing S303 as shown in FIG. 3 and any possible operation in this step), and the processor can also be used to obtain different road layers corresponding to the multiple positioning pictures. (for example, performing S304 as shown in FIG. 3 and any possible operation in this step). It should be understood that, the processor may also be used to perform other operations in the foregoing embodiments, so as to implement any map drawing method provided by the embodiments of the present application.
图17示出了的一种地图绘制装置1700的组成示意图。如图17所示,该地图绘制装置1700可以包括:处理器1701和存储器1702。该存储器1702用于存储计算机执行指令。示例性的,在一些实施例中,当该处理器1701执行该存储器1702存储的指令时,可以使得该通信装置1700执行上述实施例中任一种所示的地图绘制方法。FIG. 17 shows a schematic diagram of the composition of a map drawing apparatus 1700 . As shown in FIG. 17 , the map drawing apparatus 1700 may include: a processor 1701 and a memory 1702 . The memory 1702 is used to store computer-implemented instructions. Exemplarily, in some embodiments, when the processor 1701 executes the instructions stored in the memory 1702, the communication device 1700 can be caused to execute the map drawing method shown in any one of the foregoing embodiments.
需要说明的是,上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述,在此不再赘述。It should be noted that, all relevant contents of the steps involved in the above method embodiments can be cited in the functional description of the corresponding functional module, which will not be repeated here.
图18示出了的一种芯片***1800的组成示意图。该芯片***1800可以包括:处理器1801和通信接口1802,用于支持相关设备实现上述实施例中所涉及的功能。在一种可能的设计中,芯片***还包括存储器,用于保存终端必要的程序指令和数据。该芯片***,可以由芯片构成,也可以包含芯片和其他分立器件。需要说明的是,在本申请的一些实现方式中,该通信接口1802也可称为接口电路。FIG. 18 shows a schematic composition diagram of a chip system 1800 . The chip system 1800 may include: a processor 1801 and a communication interface 1802, which are used to support related devices to implement the functions involved in the above embodiments. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the terminal. The chip system may be composed of chips, or may include chips and other discrete devices. It should be noted that, in some implementation manners of the present application, the communication interface 1802 may also be referred to as an interface circuit.
需要说明的是,上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述,在此不再赘述。It should be noted that, all relevant contents of the steps involved in the above method embodiments can be cited in the functional description of the corresponding functional module, which will not be repeated here.
在上述实施例中的功能或动作或操作或步骤等,可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件程序实现时,可以全部或部分地以计算机程序产品的形式来实现。该计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时,全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中,或者从一个计算机可读存储 介质向另一个计算机可读存储介质传输,例如,所述计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line,DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包括一个或多个可以用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如,软盘、硬盘、磁带),光介质(例如,DVD)、或者半导体介质(例如固态硬盘(solid state disk,SSD))等。The functions or actions or operations or steps in the above embodiments may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line, DSL) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or data storage devices including one or more servers, data centers, etc. that can be integrated with the medium. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)), and the like.
尽管结合具体特征及其实施例对本申请进行了描述,显而易见的,在不脱离本申请的精神和范围的情况下,可对其进行各种修改和组合。相应地,本说明书和附图仅仅是所附权利要求所界定的本申请的示例性说明,且视为已覆盖本申请范围内的任意和所有修改、变化、组合或等同物。显然,本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样,倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内,则本申请也意图包括这些改动和变型在内。Although the application has been described in conjunction with specific features and embodiments thereof, it will be apparent that various modifications and combinations can be made therein without departing from the spirit and scope of the application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined by the appended claims, and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of this application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims (14)
- 一种地图绘制方法,其特征在于,所述方法包括:A map drawing method, characterized in that the method comprises:获取第一待绘制区域的激光点云数据,所述第一待绘制区域包括处于不同平面的第一道路和第二道路;acquiring laser point cloud data of a first area to be drawn, where the first area to be drawn includes a first road and a second road on different planes;根据所述激光点云数据,获取第一定位图片的集合和第二定位图片的集合,其中,所述集合中任意两个第一定位图片的参考高度相差不超过第一阈值;任意两个第二定位图片的参考高度相差不超过第一阈值;第一定位图片和第二定位图片的参考高度差大于第二阈值;定位图片的参考高度是采集所述定位图片对应的激光点云数据时,对应设备所在路面的高度;所述第一阈值和所述第二阈值均为正数;According to the laser point cloud data, a set of first positioning pictures and a set of second positioning pictures are obtained, wherein the difference between the reference heights of any two first positioning pictures in the set does not exceed a first threshold; any two The difference between the reference heights of the two positioning pictures does not exceed the first threshold; the reference height difference between the first positioning picture and the second positioning picture is greater than the second threshold; the reference height of the positioning picture is when the laser point cloud data corresponding to the positioning picture is collected, corresponds to the height of the road where the device is located; the first threshold and the second threshold are both positive numbers;将所述第一定位图片进行融合,获得所述第一道路对应的地图,将所述第二定位图片进行融合,获得所述第二道路对应的地图。The first positioning picture is fused to obtain a map corresponding to the first road, and the second positioning picture is fused to obtain a map corresponding to the second road.
- 根据权利要求1所述的方法,其特征在于,所述激光点云数据包括:第一位置在全局坐标系下的三维坐标信息,其中,所述第一位置为采集所述激光点云数据的位置;The method according to claim 1, wherein the laser point cloud data comprises: three-dimensional coordinate information of a first position in a global coordinate system, wherein the first position is the point where the laser point cloud data is collected. Location;在所述根据所述激光点云数据,获取第一定位图片的集合和第二定位图片的集合之前,所述方法还包括:Before acquiring the set of first positioning pictures and the set of second positioning pictures according to the laser point cloud data, the method further includes:根据所述第一位置的三维坐标信息,获取所述第一位置相对于采集所述激光点云数据时所行驶道路的相对高度信息。According to the three-dimensional coordinate information of the first position, the relative height information of the first position relative to the road on which the laser point cloud data is collected is acquired.
- 根据权利要求2所述的方法,其特征在于,所述激光点云数据还包括:用于指示所述第一位置是否为车道的第一标识。The method according to claim 2, wherein the laser point cloud data further comprises: a first sign used to indicate whether the first position is a lane.
- 根据权利要求3所述的方法,其特征在于,在所述根据所述激光点云数据,获取第一定位图片的集合和第二定位图片的集合,包括:The method according to claim 3, wherein, in the step of obtaining a set of first positioning pictures and a set of second positioning pictures according to the laser point cloud data, comprising:根据所述激光点云数据,获取多个定位图片,每个所述定位图片对应的激光点云数据的采集时间在预设范围之内,所述每个定位图片中包括的像素的像素值由所述像素对应位置的相对高度信息和所述位置对应的激光点云数据的第一标识确定;According to the laser point cloud data, a plurality of positioning pictures are obtained, the collection time of the laser point cloud data corresponding to each positioning picture is within a preset range, and the pixel values of the pixels included in each positioning picture are determined by The relative height information of the position corresponding to the pixel and the first identification of the laser point cloud data corresponding to the position are determined;将多个所述定位图片中,处于相同道路图层的定位图片进行融合,获取所述第一定位图片的集合和所述第二定位图片的集合。The positioning pictures in the same road layer among the plurality of positioning pictures are fused to obtain the set of the first positioning pictures and the set of the second positioning pictures.
- 根据权利要求4所述的方法,其特征在于,所述获取第一定位图片的集合和第二定位图片的集合,包括:The method according to claim 4, wherein the acquiring the set of the first positioning pictures and the set of the second positioning pictures comprises:确定第一图片和第二图片的相似度,并根据所述相似度,确定所述第一图片和所述第二图片是否处于相同道路图层,所述相似度用于指示所述第一图片和所述第二图片的相似程度;其中,所述第一图片和所述第二图片为多个所述定位图片中,任意两个具有相同水平覆盖区域的定位图片;Determine the similarity between the first picture and the second picture, and determine whether the first picture and the second picture are in the same road layer according to the similarity, and the similarity is used to indicate the first picture The degree of similarity with the second picture; wherein, the first picture and the second picture are a plurality of the positioning pictures, any two positioning pictures with the same horizontal coverage area;将多个所述定位图片中,处于相同道路图层的定位图片进行融合,以获取处于所述第一道路对应的第一定位图片的集合,以及所述第二道路对应的第二定位图片的集合。The positioning pictures in the same road layer among the plurality of positioning pictures are fused to obtain a set of first positioning pictures corresponding to the first road, and a set of the second positioning pictures corresponding to the second road. gather.
- 根据权利要求5所述的方法,其特征在于,所述确定第一图片和第二图片的相似度,包括:The method according to claim 5, wherein the determining the similarity between the first picture and the second picture comprises:根据所述第一图片和所述第二图片的局部特征,确定所述第一图片和所述第二图 片的第一相似度;According to the local features of the first picture and the second picture, determine the first similarity of the first picture and the second picture;所述局部特征包括以下中的一项或多项:The local features include one or more of the following:定位图片中像素的灰度平均值,所述像素的灰度方差,所述像素的灰度协方差;The grayscale mean value of the pixel in the positioning picture, the grayscale variance of the pixel, and the grayscale covariance of the pixel;所述根据所述相似度,确定所述第一图片和所述第二图片是否处于相同道路图层,包括:The determining whether the first picture and the second picture are in the same road layer according to the similarity includes:当所述第一相似度大于第一阈值时,所述第一图片和所述第二图片处于相同道路图层。When the first similarity is greater than a first threshold, the first picture and the second picture are in the same road layer.
- 根据权利要求6所述的方法,其特征在于,所述方法还包括:The method according to claim 6, wherein the method further comprises:当所述第一相似度小于所述第一阈值时,所述第一图片和所述第二图片处于不同道路图层。When the first similarity is less than the first threshold, the first picture and the second picture are in different road layers.
- 根据权利要求5-7中任一项所述的方法,其特征在于,所述确定第一定位图片和第二定位图片的相似度,包括:The method according to any one of claims 5-7, wherein the determining the similarity between the first positioning picture and the second positioning picture comprises:根据所述第一图片和所述第二图片中对应像素的相对高度信息,确定所述第一图片和所述第二图片的第二相似度;determining the second similarity between the first picture and the second picture according to the relative height information of the corresponding pixels in the first picture and the second picture;所述根据所述相似度,确定所述第一图片和所述第二图片是否处于相同道路图层,包括:The determining whether the first picture and the second picture are in the same road layer according to the similarity includes:当所述第二相似度小于第二阈值时,所述第一图片和所述第二图片处于相同道路图层。When the second similarity is less than a second threshold, the first picture and the second picture are in the same road layer.
- 根据权利要求8所述的方法,其特征在于,所述方法还包括:The method according to claim 8, wherein the method further comprises:当所述第二相似度大于所述第二阈值时,所述第一图片和所述第二图片处于不同道路图层。When the second similarity is greater than the second threshold, the first picture and the second picture are in different road layers.
- 根据权利要求8或9所述的方法,其特征在于,所述根据所述第一图片和所述第二图片中对应像素的相对高度信息,确定所述第一图片和所述第二图片的第二相似度,包括:The method according to claim 8 or 9, wherein the determining the height of the first picture and the second picture according to the relative height information of the corresponding pixels in the first picture and the second picture The second similarity, including:针对所述第一图片和所述第二图片分别执行以下操作,以获取所述第一图片对应的特征指纹,和所述第二图片对应的特征指纹:Perform the following operations on the first picture and the second picture respectively to obtain the feature fingerprint corresponding to the first picture and the feature fingerprint corresponding to the second picture:删除图片中预设行和/或预设列的像素,以获取缩小的图片,remove pixels from preset rows and/or preset columns in the picture to get a reduced picture,根据所述缩小的图片中,各个像素的相对高度的均值,对所述缩小的图片进行归一化处理,According to the mean value of the relative height of each pixel in the reduced picture, normalize the reduced picture,根据所述归一化处理后的所述缩小的图片的各个像素值,确定所述图片对应的特征指纹;According to each pixel value of the reduced picture after the normalization process, determine the feature fingerprint corresponding to the picture;根据所述第一图片的特征指纹和所述第二图片的特征指纹,确定所述第一图片和所述第二图片的第二相似度,所述第二相似度为所述第一图片的特征指纹和所述第二图片的特征指纹的汉明距离。According to the feature fingerprint of the first picture and the feature fingerprint of the second picture, a second similarity between the first picture and the second picture is determined, and the second similarity is the Hamming distance between the feature fingerprint and the feature fingerprint of the second picture.
- 一种地图绘制装置,其特征在于,所述地图绘制装置包括一个或多个处理器和一个或多个存储器;所述一个或多个存储器与所述一个或多个处理器耦合,所述一个或多个存储器存储有计算机指令;A map drawing device, characterized in that, the map drawing device comprises one or more processors and one or more memories; the one or more memories are coupled with the one or more processors, and the one or more memories storing computer instructions;当所述一个或多个处理器执行所述计算机指令时,使得所述地图绘制装置执行如权利要求1-10中任一项所述的地图绘制方法。The computer instructions, when executed by the one or more processors, cause the mapping apparatus to perform the mapping method of any one of claims 1-10.
- 一种芯片***,其特征在于,所述芯片***包括接口电路和处理器;所述接口电路和所述处理器通过线路互联;所述接口电路用于从存储器接收信号,并向所述处理器发送所述信号,所述信号包括所述存储器中存储的计算机指令;当所述处理器执行所述计算机指令时,所述芯片***执行如权利要求1-10中任一项所述的地图绘制方法。A chip system, characterized in that the chip system includes an interface circuit and a processor; the interface circuit and the processor are interconnected by lines; the interface circuit is used to receive signals from a memory and send signals to the processor sending the signal, the signal comprising computer instructions stored in the memory; when the processor executes the computer instructions, the chip system performs the mapping of any one of claims 1-10 method.
- 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质包括计算机指令,当所述计算机指令运行时,执行如权利要求1-10中任一项所述的地图绘制方法。A computer-readable storage medium, characterized in that, the computer-readable storage medium comprises computer instructions, and when the computer instructions are executed, the map drawing method according to any one of claims 1-10 is executed.
- 一种计算机程序产品,其特征在于,所述计算机程序产品中包括指令,当所述计算机程序产品在计算机上运行时,使得计算机可以根据所述指令执行如权利要求1-10中任一项所述的地图绘制方法。A computer program product, characterized in that, the computer program product includes instructions, when the computer program product is run on a computer, the computer can execute the method according to any one of claims 1-10 according to the instructions. the map drawing method described above.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010902739.7A CN114120795B (en) | 2020-09-01 | 2020-09-01 | Map drawing method and device |
CN202010902739.7 | 2020-09-01 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022048193A1 true WO2022048193A1 (en) | 2022-03-10 |
Family
ID=80360262
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/094917 WO2022048193A1 (en) | 2020-09-01 | 2021-05-20 | Map drawing method and apparatus |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN114120795B (en) |
WO (1) | WO2022048193A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114720993A (en) * | 2022-03-30 | 2022-07-08 | 上海木蚁机器人科技有限公司 | Robot positioning method, robot positioning device, electronic device, and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130230206A1 (en) * | 2012-03-01 | 2013-09-05 | Exelis, Inc. | Foliage penetration based on 4d lidar datasets |
CN108806472A (en) * | 2017-05-03 | 2018-11-13 | 腾讯科技(深圳)有限公司 | Road rendering intent, device and processing method in electronic map, device |
CN109084786A (en) * | 2018-08-09 | 2018-12-25 | 北京智行者科技有限公司 | A kind of processing method of map datum |
CN109685898A (en) * | 2018-12-25 | 2019-04-26 | 广州文远知行科技有限公司 | Point cloud data layering method and device, computer equipment and storage medium |
CN109887057A (en) * | 2019-01-30 | 2019-06-14 | 杭州飞步科技有限公司 | The method and apparatus for generating high-precision map |
CN110163900A (en) * | 2019-05-31 | 2019-08-23 | 北京百度网讯科技有限公司 | For adjusting point cloud data method and apparatus |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8798372B1 (en) * | 2012-03-07 | 2014-08-05 | Hrl Laboratories, Llc | Method for detecting bridges using lidar point cloud data |
CN105956088B (en) * | 2016-04-29 | 2019-05-21 | 百度在线网络技术(北京)有限公司 | A kind of determination method and device of road gland relationship |
CN107545833B (en) * | 2016-06-24 | 2019-09-24 | 高德信息技术有限公司 | A kind of road method for drafting and device |
CN109064506B (en) * | 2018-07-04 | 2020-03-13 | 百度在线网络技术(北京)有限公司 | High-precision map generation method and device and storage medium |
CN110160502B (en) * | 2018-10-12 | 2022-04-01 | 腾讯科技(深圳)有限公司 | Map element extraction method, device and server |
CN109544456B (en) * | 2018-11-26 | 2022-04-15 | 湖南科技大学 | Panoramic environment sensing method based on two-dimensional image and three-dimensional point cloud data fusion |
CN110210316B (en) * | 2019-05-07 | 2022-08-12 | 南京理工大学 | Traffic signal lamp digital identification method based on gray level image |
-
2020
- 2020-09-01 CN CN202010902739.7A patent/CN114120795B/en active Active
-
2021
- 2021-05-20 WO PCT/CN2021/094917 patent/WO2022048193A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130230206A1 (en) * | 2012-03-01 | 2013-09-05 | Exelis, Inc. | Foliage penetration based on 4d lidar datasets |
CN108806472A (en) * | 2017-05-03 | 2018-11-13 | 腾讯科技(深圳)有限公司 | Road rendering intent, device and processing method in electronic map, device |
CN109084786A (en) * | 2018-08-09 | 2018-12-25 | 北京智行者科技有限公司 | A kind of processing method of map datum |
CN109685898A (en) * | 2018-12-25 | 2019-04-26 | 广州文远知行科技有限公司 | Point cloud data layering method and device, computer equipment and storage medium |
CN109887057A (en) * | 2019-01-30 | 2019-06-14 | 杭州飞步科技有限公司 | The method and apparatus for generating high-precision map |
CN110163900A (en) * | 2019-05-31 | 2019-08-23 | 北京百度网讯科技有限公司 | For adjusting point cloud data method and apparatus |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114720993A (en) * | 2022-03-30 | 2022-07-08 | 上海木蚁机器人科技有限公司 | Robot positioning method, robot positioning device, electronic device, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN114120795A (en) | 2022-03-01 |
CN114120795B (en) | 2023-03-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10964054B2 (en) | Method and device for positioning | |
WO2021233029A1 (en) | Simultaneous localization and mapping method, device, system and storage medium | |
US11971726B2 (en) | Method of constructing indoor two-dimensional semantic map with wall corner as critical feature based on robot platform | |
CN110163930B (en) | Lane line generation method, device, equipment, system and readable storage medium | |
CN113657224B (en) | Method, device and equipment for determining object state in vehicle-road coordination | |
US11625851B2 (en) | Geographic object detection apparatus and geographic object detection method | |
US20230215187A1 (en) | Target detection method based on monocular image | |
US20220343758A1 (en) | Data Transmission Method and Apparatus | |
WO2022206517A1 (en) | Target detection method and apparatus | |
CN112991534B (en) | Indoor semantic map construction method and system based on multi-granularity object model | |
CN114325634A (en) | Method for extracting passable area in high-robustness field environment based on laser radar | |
CN115147328A (en) | Three-dimensional target detection method and device | |
WO2022048193A1 (en) | Map drawing method and apparatus | |
CN115147809A (en) | Obstacle detection method, device, equipment and storage medium | |
CN114387576A (en) | Lane line identification method, system, medium, device and information processing terminal | |
CN111833443A (en) | Landmark position reconstruction in autonomous machine applications | |
WO2021189420A1 (en) | Data processing method and device | |
CN116736259A (en) | Laser point cloud coordinate calibration method and device for tower crane automatic driving | |
CN115457780B (en) | Vehicle flow and velocity automatic measuring and calculating method and system based on priori knowledge set | |
CN114792343B (en) | Calibration method of image acquisition equipment, method and device for acquiring image data | |
CN116385994A (en) | Three-dimensional road route extraction method and related equipment | |
CN113222025B (en) | Feasible region label generation method based on laser radar | |
CN115618602A (en) | Lane-level scene simulation method and system | |
CN114565906A (en) | Obstacle detection method, obstacle detection device, electronic device, and storage medium | |
CN118172423B (en) | Sequential point cloud data pavement element labeling method and device and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21863263 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21863263 Country of ref document: EP Kind code of ref document: A1 |