WO2022021209A9 - 电子地图生成方法、装置、计算机设备和存储介质 - Google Patents

电子地图生成方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2022021209A9
WO2022021209A9 PCT/CN2020/105790 CN2020105790W WO2022021209A9 WO 2022021209 A9 WO2022021209 A9 WO 2022021209A9 CN 2020105790 W CN2020105790 W CN 2020105790W WO 2022021209 A9 WO2022021209 A9 WO 2022021209A9
Authority
WO
WIPO (PCT)
Prior art keywords
track unit
track
unit
path
current
Prior art date
Application number
PCT/CN2020/105790
Other languages
English (en)
French (fr)
Other versions
WO2022021209A1 (zh
Inventor
李煊
Original Assignee
深圳元戎启行科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳元戎启行科技有限公司 filed Critical 深圳元戎启行科技有限公司
Priority to PCT/CN2020/105790 priority Critical patent/WO2022021209A1/zh
Priority to CN202080099129.6A priority patent/CN115668333A/zh
Publication of WO2022021209A1 publication Critical patent/WO2022021209A1/zh
Publication of WO2022021209A9 publication Critical patent/WO2022021209A9/zh

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

Definitions

  • the present application relates to a method, apparatus, computer equipment and storage medium for generating an electronic map.
  • the existing electronic map generation methods can usually be divided into several steps, such as data collection, data preprocessing, map generation, semantic information extraction, and data post-processing.
  • the preprocessing process of the three-dimensional data collected from the multi-level driving space is not accurate, thus affecting the accuracy of the electronic map.
  • an electronic map method, apparatus, computer device and storage medium are provided.
  • a method for generating an electronic map comprising:
  • the path track includes at least one track unit; the track unit includes collection time and spatial position information; the path track is obtained by sorting the at least one track unit according to the collection time;
  • the arrangement order determine the current sequence track unit in the path track, and at least one previous sequence track unit located before the current sequence track unit;
  • the spatial position information of the current sequence track unit and the spatial position information of the previous sequence track unit filter out the matching track unit that matches the current sequence track unit from the at least one previous sequence track unit;
  • next sequential track unit is determined, and the next sequential track unit is used as the current sequential track unit to iterate until the last sequential track unit;
  • An electronic map is generated based on the track units marked with hierarchical information.
  • a device for generating an electronic map characterized in that the device comprises:
  • a path trajectory acquisition module configured to acquire a path trajectory;
  • the path trajectory includes at least one trajectory unit;
  • the trajectory unit includes acquisition time and spatial position information;
  • the path trajectory is obtained by sorting the at least one trajectory unit according to the acquisition time;
  • the hierarchical marking module is used to determine, according to the arrangement order, the current sequential track unit in the path track, and at least one preceding track unit located before the current sequential track unit; according to the spatial position information of the current sequential track unit and the preceding track unit.
  • the spatial position information of the sequence track unit, and the matching track unit that matches the current sequence track unit is screened out from the at least one pre-order track unit; the level information of the matching track unit is obtained, and based on the level information
  • the current sequence track unit is level marked; according to the arrangement order, the next sequence track unit is determined, and the next sequence track unit is used as the current sequence track unit to iterate until the last sequence track unit;
  • An electronic map generation module is used to generate an electronic map based on the track units marked with hierarchical information.
  • a computer device comprising a memory and one or more processors, the memory having computer-readable instructions stored therein, the computer-readable instructions, when executed by the processor, cause the one or more processors to execute The following steps:
  • the path track includes at least one track unit; the track unit includes collection time and spatial position information; the path track is obtained by sorting the at least one track unit according to the collection time;
  • the arrangement order determine the current sequence track unit in the path track, and at least one previous sequence track unit located before the current sequence track unit;
  • the spatial position information of the current sequence track unit and the spatial position information of the previous sequence track unit filter out the matching track unit that matches the current sequence track unit from the at least one previous sequence track unit;
  • next sequential track unit is determined, and the next sequential track unit is used as the current sequential track unit to iterate until the last sequential track unit;
  • An electronic map is generated based on the track units marked with hierarchical information.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the following steps:
  • the path track includes at least one track unit; the track unit includes collection time and spatial position information; the path track is obtained by sorting the at least one track unit according to the collection time;
  • the arrangement order determine the current sequence track unit in the path track, and at least one previous sequence track unit located before the current sequence track unit;
  • the spatial position information of the current sequence track unit and the spatial position information of the previous sequence track unit filter out the matching track unit that matches the current sequence track unit from the at least one previous sequence track unit;
  • next sequential track unit is determined, and the next sequential track unit is used as the current sequential track unit to iterate until the last sequential track unit;
  • An electronic map is generated based on the track units marked with hierarchical information.
  • Fig. 1 is the application scene diagram of the electronic map generation method in one embodiment
  • FIG. 2 is a schematic flowchart of a method for generating an electronic map in one embodiment
  • 3A is a bird's-eye view schematic diagram of a spatial coordinate system in one embodiment
  • 3B is a three-dimensional schematic diagram of a space coordinate system in one embodiment
  • FIG. 4 is a schematic flowchart of generating an electronic map based on a grid map according to one embodiment
  • FIG. 5 is a block diagram of an electronic map generating apparatus in one embodiment
  • FIG. 6 is a block diagram of an electronic map generating apparatus in another embodiment
  • Figure 7 is a block diagram of a computer device in one embodiment.
  • the electronic map generation method provided in this application can be applied to the application environment shown in FIG. 1 .
  • the surveying and mapping vehicle 102 communicates with the server 104 through the network.
  • the surveying and mapping vehicle 102 refers to a vehicle that is installed with various sensors and collects field data based on the various sensors.
  • the surveying and mapping vehicle 102 collects track units, and sends the collected track units to the server 104 .
  • the server 104 preprocesses the track units, and generates an electronic map based on the preprocessed track units.
  • the server 104 may be implemented by an independent server or a server cluster composed of multiple servers.
  • a method for generating an electronic map is provided, and the method is applied to the server in FIG. 1 as an example for description, including the following steps:
  • S202 Acquire a path track; the path track includes at least one track unit; the track unit includes collection time and spatial position information; the path track is obtained by sorting the at least one track unit according to the collection time.
  • the track unit refers to the collection of data collected by various sensors installed in the surveying and mapping vehicle.
  • the trajectory unit includes spatial position information, a three-dimensional point cloud image, a collection time for collecting the spatial position information and the three-dimensional point cloud information.
  • the spatial position information refers to the position coordinates of the surveying and mapping vehicle in the Cartesian coordinate system
  • the three-dimensional point cloud image refers to the point cloud data collected by the laser sensor erected on the surveying and mapping vehicle
  • the acquisition time refers to the space obtained by the surveying and mapping vehicle. location information and the time point of the 3D point cloud image. It is easy to understand that the acquisition time of the spatial position information in the same trajectory unit is consistent with the acquisition time of the three-dimensional point cloud image.
  • a laser sensor and a GPS (Global Positioning System) positioning system can be set up on the surveying and mapping vehicle in advance, and the laser sensor transmits a detection signal to the driving area according to the preset collection frequency, and reflects the signal back from the objects in the driving area. Compare with the detection signal to obtain surrounding environmental data, and generate a three-dimensional point cloud image based on the environmental data; and determine the latitude and longitude coordinates and altitude of the current position by the GPS positioning system according to the same preset collection frequency, and altitude are converted to spatial location information.
  • GPS Global Positioning System
  • the surveying and mapping vehicle When the surveying and mapping vehicle receives the data collection instruction sent by the server, the surveying and mapping vehicle controls the GPS positioning system to determine the latitude and longitude coordinates and altitude of the current position, and establishes a spatial coordinate system with the center of the laser sensor erected on the surveying and mapping vehicle as the origin. The latitude and longitude coordinates and altitude of the location are converted to three-dimensional spatial location coordinates in the spatial coordinate system.
  • the established spatial coordinate system takes the center of the laser sensor erected on the surveying and mapping vehicle as the origin; the horizontal plane horizontal to the ground is the reference plane (that is, the horizontal plane horizontal to the ground is XOY plane); the axis horizontal to the direction of movement of the surveying and mapping vehicle is the Y axis; the axis in the datum plane, passing through the origin and perpendicular to the Y axis is the X axis; the axis passing through the origin and perpendicular to the datum plane is the Z axis.
  • FIG. 3A is a schematic bird's-eye view of a spatial coordinate system in one embodiment.
  • FIG. 3A is a schematic bird's-eye view of a spatial coordinate system in one embodiment.
  • 3B is a three-dimensional schematic diagram of a space coordinate system in one embodiment. It is easy to understand that before receiving the data collection stop instruction sent by the server, the GPS positioning system converts the latitude and longitude coordinates and the altitude based on the spatial coordinates established when the data collection instruction is received.
  • the surveying and mapping vehicle controls the laser sensor to collect the surrounding environment data, and generates a 3D point cloud image based on the surrounding environment data.
  • the surveying and mapping vehicle integrates the collection time of the 3D point cloud image, the spatial position information, the 3D point cloud image and the spatial position information to obtain the trajectory unit, which will be collected in the collection time.
  • the at least one track unit collected during the period is sent to the server.
  • the server receives the at least one trajectory unit collected during the collection period, and sorts the received at least one trajectory unit based on the collection time in the trajectory unit to obtain the path trajectory. For example, when the surveying and mapping vehicle passes through a multi-storey parking lot during the collection period, the surveying and mapping vehicle collects its spatial position information in the multi-storey parking lot and the 3D point cloud image of the multi-storey parking lot according to the preset collection frequency, and then uses Spatial location information, 3D point cloud images and acquisition time are combined into trajectory units. When the collection is over, the surveying and mapping vehicle sorts the multiple trajectory units in ascending order of collection time, and obtains the path trajectory in the multi-storey parking lot.
  • the spatial positioning information and the three-dimensional point cloud images whose acquisition time difference is within the preset difference range can also be combined into the same trajectory unit, and
  • the acquisition time of the spatial positioning information or the acquisition time of the 3D point cloud image is taken as the acquisition time in the trajectory unit.
  • the collection time of the spatial positioning information of A is a
  • the collection time of the three-dimensional point cloud image of B is b
  • the time difference between a and b is less than the preset time threshold
  • the spatial location information is often inaccurate, resulting in ghost images in the generated electronic map.
  • the direct method, optical flow method, Kalman filter, graph optimization and other calculations can be used to optimize the spatial position information collected by the GPS positioning system.
  • S204 determine the current sequential track unit in the path track, and at least one pre-sequence track unit located before the current sequential track unit.
  • the previous sequence track unit refers to the data whose collection time is before the collection time of the current sequence track unit.
  • the server traverses the track unit except the first track unit in the path track, takes the track unit in the current traversal order as the current sequence track unit, and uses the path track in the path track.
  • the data located before the current sequential track unit is used as the previous sequential track unit, and so on until the last sequential track unit.
  • the server takes the track unit in the second order in the path track as the current sequence track unit, and the first sequence track unit as the pre-order track unit.
  • the server takes the third order track unit as the current order track unit, and uses both the first order track unit and the second order track unit as the pre-order track unit.
  • the server extracts the spatial position information in the current sequence track unit, and extracts the spatial position information in each pre-order track unit, and compares the spatial position information in the current sequence track unit with the spatial position information in each pre-order track unit respectively.
  • the location information is used for information matching, and when there is a successfully matched pre-order trajectory unit in at least one pre-sequence point data, the server uses the successfully matched pre-order trajectory unit as a matching trajectory unit.
  • the distance between the X-axis and the distance between the Y-axis are both smaller than the preset distance threshold, and the distance between the Z-axis is greater than the preset distance threshold
  • the pre-order trajectory unit is used as the matching trajectory unit.
  • the spatial position information includes horizontal plane coordinates and vertical coordinates; according to the spatial position information of the current sequence track unit and the spatial position information of the previous sequence track unit, screen out at least one previous sequence track unit and the current sequence track unit.
  • the matching track unit matching the track unit includes: determining the horizontal distance between the current sequence track unit and different pre-order track units according to the horizontal plane coordinates; according to the vertical coordinates, determining the current sequence track unit and the different pre-order track units Vertical spacing; according to the horizontal spacing and vertical spacing, filter out the matching track unit that matches the current sequence track unit from at least one previous sequence track unit.
  • the horizontal plane refers to the XOY plane in the space coordinate system
  • X refers to the X axis in the space coordinate system
  • Y refers to the Y axis in the space coordinate system
  • O refers to the origin in the space coordinate system.
  • the horizontal plane coordinates refer to the position coordinates of the surveying and mapping vehicle in the XOY plane.
  • the vertical coordinate refers to the Z-axis coordinate in the space coordinate system. For example, when the spatial position coordinates of the surveying vehicle in the spatial coordinate system are (x, y, z), (x, y) is the horizontal coordinate; (z) is the vertical coordinate.
  • the server extracts the horizontal plane coordinates in the current sequential trajectory unit, and extracts the horizontal plane coordinates in each pre-order trajectory unit respectively, and extracts the horizontal plane coordinates according to the preset formula Calculates the horizontal spacing between the current sequential track unit and different preceding track units.
  • x a refers to the X-axis coordinate value in the horizontal plane coordinates of the current sequential trajectory unit, that is, the X-axis coordinate value in the spatial information in the current sequential trajectory unit
  • x b refers to the horizontal plane coordinate value of the previous sequential trajectory unit X Axis coordinate value, that is, the X-axis coordinate value in the spatial information in the previous sequence track unit
  • y a The Y-axis coordinate value in the horizontal plane coordinates of the current sequence track unit, that is, the X-axis coordinate value in the spatial information in the current sequence track unit
  • y b refers to the Y-axis coordinate value in the horizontal plane coordinates of the preceding track unit, that is, the Y-axis coordinate value in the spatial information in the preceding track unit.
  • the horizontal coordinates of the current sequence track unit are (x 1 , y 1 ), the horizontal coordinates of the pre-order track unit A are (x 2 , y 2 ), and the horizontal coordinates of the pre-order track unit B are (x 3 , y ) 3 ), the current sequence track unit and A
  • the server extracts the vertical coordinates in the current sequential trajectory unit, and extracts the vertical coordinates in each pre-order trajectory unit respectively, and calculates the current sequential trajectory unit and different pre-order trajectories according to the preset formula
  • za refers to the vertical coordinate of the current sequence track unit, that is, the Z-axis coordinate value in the spatial information in the current sequence track unit
  • z b refers to the vertical coordinate value of the previous sequence track unit, that is, the previous sequence track unit. Z coordinate value in spatial information.
  • the current sequential track unit and the vertical coordinate of the preceding track unit of B are (z 3 ).
  • the vertical spacing between the previous track units of A is
  • the vertical spacing between the current sequential track unit and the previous track unit of B is
  • the server selects a matching track from at least one of the preceding track units according to the horizontal spacing between the current sequential track unit and each preceding track unit, and the vertical spacing between the current sequential track unit and each preceding track unit. unit.
  • screening out a matching track unit that matches the current sequential track unit from at least one preceding track unit including: selecting a preceding track unit whose horizontal spacing and vertical spacing both meet a preset condition.
  • the sequential track unit is used as the matching track unit.
  • the server determines that the horizontal spacing is smaller than the preset horizontal threshold, and the preceding track units whose vertical spacing is greater than the preset vertical threshold are used as matching track units.
  • the horizontal distance between the current track unit and the previous track unit is smaller than the preset horizontal threshold, and the vertical distance is greater than the preset vertical threshold, it indicates that the surveying vehicle is in the same position between the collection time of the current sequence track unit and the collection time of the matching track unit.
  • the same driving space, such as in the same parking lot, or in the same viaduct, and the height of the surveying and mapping vehicle at the acquisition time of the current sequential track unit is inconsistent with the height at the acquisition time of the matching track unit.
  • the acquisition time of the sequential track unit is at the upper level of the viaduct, and the collection time of the matching track unit is at the lower level of the viaduct.
  • the server takes the A pre-order track unit as the matching track unit that matches the current sequence track unit.
  • the preset horizontal threshold and the preset vertical threshold can be freely set according to requirements.
  • the preset horizontal threshold can be set according to the lane width in the driving path
  • the preset vertical threshold can be set according to the floor height in the driving space. threshold.
  • each track unit By determining the spatial position information of each track unit, it is possible to determine, based on the spatial position information, matching track units that are in the same driving space as the current sequential track unit but have different height values, so that the current sequence track unit can be subsequently determined based on the level information of the matching track unit.
  • the level information is marked.
  • S208 Acquire hierarchical information of the matching track unit, and perform hierarchical marking on the current sequential track unit based on the hierarchical information.
  • the level information refers to the information used to identify the level of the current position.
  • the level information of the current surveying and mapping vehicle can be determined based on the level information, and the current mapping vehicle can also be determined based on the level information.
  • the server sets initial level information for all track units in the path track, that is, the server sets the level information of all track units in the path track to the same initial value. All track units in the track have their level flags set to 0.
  • the server modifies the initial level information of the current sequence track unit based on the level information of the matched track unit.
  • the server when it is determined based on the spatial position information that the Z-axis coordinate value of the current sequential track unit is greater than the Z-axis coordinate value of the matching track unit, the server increases the level information of the matching track unit, and adds the increased level information as the level information of the current sequential track unit.
  • the surveying and mapping vehicle collects data for a multi-storey parking lot
  • the driving space is higher than the formal space when the track unit is matched, which means that the parking lot floor where the mapping vehicle is at the current track unit is higher than the parking lot floor when the track unit is matched.
  • the server obtains the matching The level identification of the track unit, and correspondingly increase the obtained level identification, and use the increased level identification as the level identification of the current sequential track path.
  • the server reduces the level information of the matching track unit, and uses the reduced level information as the level information of the current sequential track unit.
  • S210 Determine the next sequential track unit according to the arrangement order, and use the next sequential track unit as the current sequential track unit to iterate until the last sequential track unit.
  • the server takes the track unit in the path track that is adjacent to the current sequential track unit and located after the current sequential track unit as the next sequential track unit, and assigns the next sequential track unit to the next sequential track unit.
  • the track unit is used as the current sequential track unit, returning to the step of determining at least one preceding track unit located before the current sequential track unit until the last sequential track unit.
  • the server groups the track units in the path track according to the level information, and groups the track units with the same level information into one group to obtain at least one group of track unit groups.
  • the server obtains a grid map template corresponding to each track unit group, and projects the track units in the track unit group into the corresponding grid map template, thereby rendering each grid image in the grid map template. Get a grid map.
  • the grid map is composed of grid images of the same size with M rows and N columns, where M and N are positive integers; the relevant information of the grid map includes: the number of rows and columns of the grid map, the The unit pixel length, the horizontal and vertical pixels of each grid image in the grid map, where the unit pixel length of the grid map is used to represent the physical length represented by one pixel in the grid map, for example, the The meaning of the unit pixel length of 3 cm is that the physical length represented by each pixel in the grid map is 3 cm.
  • a grid map template refers to a grid map in which grid images are not rendered.
  • a grid map template may be a grid map composed of multiple blank grid images.
  • the server inputs the rendered grid map into the trained semantic recognition model, and extracts semantic information in the grid map through the semantic recognition model.
  • the semantic information refers to the traffic information in the grid map.
  • Traffic information includes, but is not limited to, at least one of the following: road shape information, road gradient information, road curvature information, road direction information, lane width information, crash barrier information, road edge information, lane line information, and diversion line information, etc.
  • the server loads the semantic information into the corresponding grid map, and obtains an electronic map corresponding to each hierarchical track unit group. For example, when the server generates grid map A corresponding to track unit group A, and generates grid map B corresponding to track unit group B, the server extracts semantic information A in grid map A based on the semantic recognition model , and extract the semantic information B in the B grid map, load the semantic information A into the A grid map, obtain the electronic map A corresponding to the A track unit group, and load the semantic information B into the B grid map , obtain the electronic map B corresponding to the B track unit group.
  • the server combines the electronic maps corresponding to each track unit group to obtain the electronic map in the multi-level space.
  • the current sequence track unit and the previous track unit can be determined according to the arrangement order of the track units in the path track; by determining the current sequence track unit and the previous track unit, the current sequence track unit can be obtained.
  • the spatial position information of the trajectory unit and the spatial position information of the pre-order trajectory unit, and the matching trajectory unit that matches the current sequence trajectory unit is selected from the pre-order trajectory according to the determined spatial position information; by determining the matching matching trajectory unit , the current sequential track unit can be hierarchically marked according to the hierarchical information of the matching track unit, so that the track units in the path track are iterated, so as to realize the hierarchical marking of each track unit in the path track, and then the hierarchical information can be marked according to the marked level information.
  • the trajectory unit generates the corresponding electronic map. Since the present application adds hierarchical information to the track units in the path track, in the process of generating the electronic map, when reducing the dimension of the three-dimensional track unit collected for the multi-level driving space into two-dimensional data, you can After dimensionality reduction, overlapping trajectory units are distinguished, thereby reducing the probability of overlapping of generated electronic maps due to overlapping trajectory units, and improving the accuracy of electronic maps.
  • acquiring hierarchical information of the matching track unit, and performing hierarchical marking on the current sequential track unit based on the hierarchical information includes: when there are at least two matching track units, according to the spatial position information of the matching track unit, from the at least two track units The target track unit is selected from the matching track unit; the level information of the target track unit is obtained; the current sequence track unit is marked with a level according to the level information of the target track unit.
  • the server when the matching track units are obtained, the server counts the number of obtained matching track units. When there are two or more matching track units matching the current sequence track unit, the server extracts the vertical coordinates in each matching track unit, and calculates the absolute value of each vertical coordinate, and assigns the matching track with the largest absolute value to the matching track unit. unit as the target trajectory unit. The server acquires the level information of the target track unit, and marks the current sequence track unit according to the level information of the target track unit.
  • the target track unit is selected from multiple matching track units, so that the server can directly adjust the level information of the current sequential track unit according to the level information of the target track unit. In this way, the level information of the current sequential track unit is improved. determinate efficiency.
  • the spatial position information includes vertical coordinates; the hierarchical information is a hierarchical identification; and the hierarchical marking of the current sequential track point according to the hierarchical information of the target track point includes: when it is determined that the vertical coordinate of the current sequential track unit is greater than the target track unit When the vertical coordinates of , increase the level identifier of the target track unit; perform level mark on the current sequence track unit based on the increased level identifier.
  • the server extracts the vertical coordinates in the spatial position information in the target track unit, and extracts the vertical coordinates in the spatial position information in the current sequence track unit, where the vertical coordinates of the current sequence track unit are greater than the vertical coordinates of the target track unit
  • the server obtains the level identifier of the target track unit and a preset increase amplitude, increases the level identifier of the target track unit based on the increase amplitude, and uses the increased level identifier as the level identifier of the current sequential track unit. For example, when the level identification of the target track unit is l 1 and the increase amplitude is l, the level identification l 2 of the current sequential track unit is l 1 +l, and the level identification of the target track unit remains unchanged as l 1 .
  • the vertical coordinate of the current sequential track unit when the vertical coordinate of the current sequential track unit is greater than the vertical coordinate of the target track unit, it indicates that the spatial position of the surveying and mapping vehicle is higher than that under the collection time of the target track unit under the collection time of the current sequential track unit.
  • the hierarchical identification of the current sequential track unit can be obtained quickly by simply increasing the level identification of the target track unit, which greatly improves the efficiency of determining the level identification of the current sequential track unit.
  • the above-mentioned electronic map generation method further includes: when it is determined that the vertical coordinate of the current sequential track unit is smaller than the vertical coordinate of the target track unit, lowering the level identifier of the target track unit; using the reduced level identifier as the current sequential track The unit's level tag.
  • the server extracts the vertical coordinates in the spatial position information in the target track unit, and extracts the vertical coordinates in the spatial position information in the current sequence track unit, where the vertical coordinates of the current sequence track unit are smaller than the vertical coordinates of the target track unit , it indicates that the spatial position of the surveying and mapping vehicle is lower than the spatial position at the acquisition time of the target trajectory unit under the acquisition time of the current sequential trajectory unit.
  • the server obtains the level identifier of the target track unit and the preset reduction amplitude, reduces the level identifier of the target track unit based on the reduction amplitude, and uses the reduced level identifier as the level identifier of the current sequential track unit. For example, when the level identification of the target track unit is l 1 and the reduction amplitude is l, the level identification l 2 of the current sequential track unit is l 1 -l, and the level identification of the target track unit remains unchanged as l 1 .
  • generating an electronic map based on track units marked with hierarchical information includes:
  • S404 Group the track units in the path track according to the hierarchical information to obtain at least one track unit group.
  • S408 Generate an electronic map according to the grid map corresponding to each track unit group.
  • the trajectory unit includes a three-dimensional point cloud image collected by a laser sensor for the driving area.
  • a 3D point cloud image includes 3D coordinates and reflection values of 3D point data.
  • the three-dimensional coordinates of the three-dimensional point data refer to the coordinate values in the space coordinate system of the reflection points in the driving area obtained by laser scanning.
  • Reflectance values include color values and/or luminance values.
  • the reflection value can be the three primary colors in the color mode; the reflection value can also be the hue value, saturation value and lightness value in the color mode; the reflection value can also be the hue value, saturation value and brightness value in the color mode.
  • the parameter information includes the number of rows and columns of the grid map, the unit pixel length of the grid map, and the horizontal and vertical pixels of each grid image in the grid map.
  • the server groups each track unit in the path track, regards the track units with the same level information as a track unit group, and obtains preset parameter information for generating a grid map.
  • the server traverses each track unit group, generates a grid map corresponding to each track unit group according to the parameter information, and synthesizes the grid map corresponding to each track unit group to obtain the grid map of the path track.
  • the server inputs the grid map corresponding to the path track into the pre-trained semantic recognition model, and generates an electronic map of the path track based on the semantic recognition model.
  • the semantic recognition model refers to a machine learning model that can extract semantic information from grid maps.
  • the server determines the track unit group in the current traversal order, and generates parameter information of the grid map, and generates a grid map template corresponding to the track unit group in the current traversal order according to the parameter information.
  • the server traverses each track unit in the track unit group in the current traversal sequence, projects the track unit into the grid map template, and obtains a grid map corresponding to the current traversal sequence.
  • the server determines the trajectory unit of the current traversal order in the trajectory unit group of the current traversal order, and determines each 3D point according to the parameter information and the 3D coordinates of each 3D point data in the 3D point cloud image in the trajectory unit of the current traversal order
  • the location information of the data in the grid map template For example, when the 3D coordinates of the current 3D point data are (x 4 , y 4 , z 4 ), the corresponding location information in the grid map template is M 4 row N 4 List.
  • the server determines the grid image in the grid map template that has the same location information as the current 3D point data, and renders the grid image with the same location information according to the reflection value in the current 3D point data, so as to project the 3D point data to two Dimension grid map template.
  • the server sequentially projects the 3D point data in the 3D point cloud image to the grid map template, so as to realize the projection of the current traversal sequence track unit to the grid map template.
  • the track units in the path track are grouped by hierarchical information, and the track units with the same level information can be divided into the same group, so that the track units of the same level can be projected into the same grid map template, so , which can reduce the probability of overlapping of the trajectory units due to the loss of height information when projecting the trajectory units collected for the multi-level driving space onto the two-dimensional grid map, thereby improving the accuracy of the grid map.
  • generating an electronic map according to the grid map corresponding to each track unit group includes: respectively inputting the grid map corresponding to each track unit group into a pre-trained semantic recognition model, and obtaining each Semantic information corresponding to each grid map; an electronic map is generated according to the corresponding semantic information of each grid map.
  • the server pre-stores the trained semantic recognition model.
  • the server inputs the grid map into the semantic recognition model, extracts the semantic information in the grid map based on the semantic recognition model, and loads the semantic information into the corresponding grid map , the corresponding electronic map of each track unit group is obtained.
  • the server when the server extracts the lane width information from the grid map corresponding to the A track unit group based on the semantic recognition model, and extracts the road edge information from the grid map corresponding to the B track unit group, the server converts the lane width The information is loaded into the grid map corresponding to the A track unit group, and the electronic map corresponding to the A track unit group is obtained; the road edge information is loaded into the grid map corresponding to the B track unit, and the corresponding to the B track unit is obtained. digital map. Further, the server integrates the electronic maps loaded with semantic information to obtain the electronic map of the path track.
  • the electronic map that can be generated based on the semantic information is more accurate.
  • the above-mentioned method for generating an electronic map includes: when more than one path track is collected during the collection period, determining the collection time of each track unit in the more than one path track; The collection time of each track unit determines the time offset corresponding to more than one path track; according to the time offset, the collection time of the track point in the corresponding path track is modified; according to the modified collection time, for more than one track
  • the path trajectories are merged to obtain the merged path trajectories.
  • the server Since the track units in the path track need to be sorted according to the acquisition time before the level marking is performed on the track units.
  • the server receives multiple path trajectories collected by multiple surveying and mapping vehicles in the same collection time period, the multiple path trajectories in the multiple path trajectories will overlap in the collection time, resulting in the merging of the multiple path trajectories. , the spatial position of the merged path trajectory will produce jumps.
  • the acquisition time of the trajectory units collected by different surveying and mapping vehicles can be modified correspondingly.
  • the server when receiving path trajectories sent by multiple surveying and mapping vehicles within the collection period, the server progressively sorts the multiple path trajectories to obtain a trajectory sequence.
  • the way of sorting the path trajectories can be freely set according to requirements, for example, sorting the path trajectories according to the time point when the path trajectories are received.
  • multiple surveying and mapping vehicles refer to two or more surveying and mapping vehicles; multiple path trajectories refer to two or more path trajectories.
  • the server traverses the path tracks in the track sequence according to the arrangement order, and determines the current traversal sequence path track and the pre-order path track located before the current traversal sequence path track.
  • the server determines the maximum acquisition time of the track unit in each pre-order path track, and determines the time offset of the current traversal sequence path track according to the maximum collection time of the track unit in each pre-order path track. The corresponding time offsets of each path track. The server modifies the acquisition time of the corresponding path trajectory based on the time offset.
  • the server sorts the track units in the multiple path tracks according to the modified collection time, so as to combine more than one path track to obtain a combined path track.
  • the time offset corresponding to each path track is determined, and the acquisition time of the track unit is modified based on the time offset, so that the combined path track is continuous in time, so that subsequent time-continuous
  • the path trajectory generates the corresponding electronic map.
  • determining the time offsets corresponding to each of the more than one path trajectory according to the acquisition time of each trajectory point in the more than one path trajectory includes: according to the acquisition time of the trajectory units in the more than one path trajectory , sort more than one path trajectory to obtain a trajectory sequence; traverse the trajectory sequence according to the sorting order; when there is at least one pre-order trajectory path located before the current traversal sequence trajectory path in the trajectory sequence, obtain each pre-order trajectory The maximum acquisition time in the path; according to the maximum acquisition time, the time offset of the trajectory path in the current traversal order is determined.
  • the server obtains the minimum collection time in each path track, and sorts the more than one path track according to the minimum collection time to obtain a track sequence. For example, the server sorts more than one path trajectory in ascending order of the minimum acquisition time to obtain a trajectory sequence.
  • the server traverses the path tracks in the track sequence in the sorted order.
  • the server determines the path trajectories of the current traversal order and all the pre-order path trajectories before the current traversal order, and adds the maximum collection time in each pre-order path trajectories to obtain the time offset of the current traversal order.
  • the server After obtaining the time offset of each path track, the server correspondingly modifies the acquisition time of the track unit in the path track according to the time offset. It is easy to understand that when the current traversal sequence path track does not have a pre-order path track, the current traversal sequence path track does not have a time offset.
  • the server when the server receives path track A, path B, and path track C within 0 to 10 seconds, the acquisition time of each track unit in path track A is [1, 3, 5, 7], and the tracking time in path track B is [1,3,5,7].
  • the collection time of each track unit is [2, 4, 6, 8], and the collection time of each track unit in path C is [3, 5, 7, 9], the server will follow the minimum value of the collection time in each path track. Sort the A-path trajectory, B-path trajectory and C-path trajectory to obtain a trajectory sequence.
  • the server takes the maximum collection time 7 in the path A path as the time offset of the path trace; superimposes the maximum collection time 7 in the A path trace and the maximum collection time 8 in the B path trace to obtain the time offset of the C path trace shift.
  • the server increases the acquisition time of the trajectory units in the B path trajectory by 7 to obtain the modified acquisition time [9, 11, 13, 15]; and increases the acquisition time of the trajectory units in the C path trajectory by 15 to obtain the modified acquisition time. later acquisition times [18, 20, 22, 24].
  • the server merges the path A, path B and path C according to the modified collection time, so that the collection time of each track unit in the merged path track is [1, 3, 5, 7, 9, 11, 13, 15, 18, 20, 22, 24], and then the server performs data processing on the merged path track based on the electronic map generation method to obtain an electronic map corresponding to the merged path track.
  • the acquisition time of the path trajectory can be modified based on the time offset subsequently, and then the path trajectory can be merged based on the modified acquisition time.
  • the server traverses the trajectory sequence according to the sorting order.
  • the server determines the current traversal sequence path track, and a pre-order path track adjacent to the current traversal sequence path track and located before the current traversal sequence path track, and uses the maximum collection time in the pre-order path track as the current traversal sequence path track.
  • the time offset is modified, and based on the determined time offset, the acquisition time of each track unit in the current traversal sequence path track is modified.
  • the server determines the next sequential path track according to the arrangement order, and uses the next sequential path track as the current traversal order path track, and returns a pre-order path track determined to be adjacent to the current traversal order path track and located before the current traversal order path track steps until the final sequential path trajectory. Further, the server combines the path trajectories after the acquisition time has been modified to obtain the combined path trajectories.
  • steps in the flowcharts of FIGS. 2 and 4 are sequentially displayed in accordance with the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders. Moreover, at least a part of the steps in FIGS. 2 and 4 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed and completed at the same time, but may be executed at different times. These sub-steps or stages are not necessarily completed at the same time. The order of execution of the steps is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of sub-steps or stages of other steps.
  • an electronic map generating apparatus 500 including: a path trajectory obtaining module 502 , a level marking module 504 , and a map generating module 506 .
  • the path trajectory acquisition module 502 is used to acquire the path trajectory; the path trajectory includes at least one trajectory unit; the trajectory unit includes acquisition time and spatial position information; the path trajectory is obtained by sorting the at least one trajectory unit according to the acquisition time;
  • the level marking module 504 is configured to determine the current sequential track unit in the path track and at least one preceding track unit located before the current sequential track unit according to the arrangement sequence; according to the spatial position information of the current sequential track unit and the preceding track unit The spatial position information of the trajectories is selected from at least one pre-order trajectory unit, and the matching trajectory unit that matches the current sequence trajectory unit is selected; the level information of the matching trajectory unit is obtained, and the current sequence trajectory unit is hierarchically marked based on the hierarchical information; according to the arrangement order , determine the next sequence track unit, and use the next sequence track unit as the current sequence track unit to iterate until the last sequence track unit;
  • the map generation module 506 is configured to generate an electronic map based on the track units marked with hierarchical information.
  • the level marking module 504 further includes a matching track unit determination module 5041, configured to determine the horizontal distance between the current sequential track unit and different preceding track units according to the coordinates of the horizontal plane; Coordinates to determine the vertical spacing between the current sequential track unit and different preceding track units; according to the horizontal spacing and vertical spacing, filter out the matching track unit that matches the current sequential track unit from at least one preceding track unit.
  • a matching track unit determination module 5041 configured to determine the horizontal distance between the current sequential track unit and different preceding track units according to the coordinates of the horizontal plane; Coordinates to determine the vertical spacing between the current sequential track unit and different preceding track units; according to the horizontal spacing and vertical spacing, filter out the matching track unit that matches the current sequential track unit from at least one preceding track unit.
  • the matching track unit determination module 5041 is further configured to use the pre-order track unit whose horizontal spacing and vertical spacing both meet the preset conditions as the matching track unit.
  • the level marking module 504 further includes a target trajectory unit determination module 5042, configured to filter out the target from the at least two matching trajectory units according to the spatial location information of the matching trajectory units when there are at least two matching trajectory units Track unit; obtain the level information of the target track unit; perform level mark on the current sequence track unit according to the level information of the target track unit.
  • a target trajectory unit determination module 5042 configured to filter out the target from the at least two matching trajectory units according to the spatial location information of the matching trajectory units when there are at least two matching trajectory units Track unit; obtain the level information of the target track unit; perform level mark on the current sequence track unit according to the level information of the target track unit.
  • the target trajectory unit determining module 5042 is further configured to increase the level identifier of the target trajectory unit when it is determined that the vertical coordinate of the current sequential trajectory unit is greater than the vertical coordinate of the target trajectory unit; and use the increased level identifier as The level marker of the current sequential track unit.
  • the target track unit determination module 5042 is further configured to lower the level identifier of the target track unit when it is determined that the vertical coordinate of the current sequence track unit is smaller than the vertical coordinate of the target track unit; and use the lowered level identifier as the current sequence The level marker for the track unit.
  • the map generation module 506 further includes a grid map generation module 5061 for acquiring parameter information for generating a grid map; grouping the track units in the path track according to the hierarchical information to obtain at least one track unit group; According to the parameter information, a grid map corresponding to each track unit group is generated; according to the grid map corresponding to each track unit group, an electronic map is generated.
  • the grid map generation module 5061 is further configured to input the grid map corresponding to each track unit group into the pre-trained semantic recognition model to obtain the corresponding semantic information of each grid map; The corresponding semantic information of each grid map generates an electronic map.
  • the electronic map generating apparatus 500 further includes a merging module 508, configured to determine the collection time of each track unit in the more than one path track when more than one path track is collected within the collection period; According to the collection time of each track unit in a path track, determine the time offset corresponding to more than one path track; modify the collection time of the track point in the corresponding path track according to the time offset; according to the modified collection time time, combine more than one path trajectory to obtain a combined path trajectory.
  • a merging module 508 configured to determine the collection time of each track unit in the more than one path track when more than one path track is collected within the collection period; According to the collection time of each track unit in a path track, determine the time offset corresponding to more than one path track; modify the collection time of the track point in the corresponding path track according to the time offset; according to the modified collection time time, combine more than one path trajectory to obtain a combined path trajectory.
  • the merging module 508 is further configured to sort more than one path trajectory according to the acquisition time of the trajectory units in the more than one path trajectory to obtain a trajectory sequence; traverse the trajectory sequence according to the sorted order; when the trajectory When there is at least one pre-order trajectory path in the sequence before the current traversal sequence trajectory path, the maximum acquisition time in each pre-order trajectory path is obtained; according to the maximum acquisition time, the time offset of the trajectory path in the current traversal order is determined.
  • a computer device in one embodiment, is provided, and the computer device can be a server, and its internal structure diagram can be as shown in FIG. 7 .
  • the computer device includes a processor, memory, a network interface, and a database connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium, an internal memory.
  • the non-volatile storage medium stores an operating system, computer readable instructions and a database.
  • the internal memory provides an environment for the execution of the operating system and computer-readable instructions in the non-volatile storage medium.
  • the database of the computer equipment is used to store electronic map data.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the computer-readable instructions when executed by a processor, implement a method of generating an electronic map.
  • FIG. 7 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.
  • a computer device comprising a memory and one or more processors, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the one or more processors implement the above method embodiments when executed. step.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions, when the computer-readable instructions are executed by one or more processors, the one or more processors implement the above method embodiments when executed. A step of.
  • Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Road (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • Educational Technology (AREA)
  • General Physics & Mathematics (AREA)
  • Image Analysis (AREA)
  • Instructional Devices (AREA)

Abstract

一种电子地图生成方法、装置、计算机设备以及存储介质，包括：获取路径轨迹；按照排列顺序，确定路径轨迹中的当前顺序轨迹单元，以及位于当前顺序轨迹单元之前的至少一个前序轨迹单元；根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元；获取匹配轨迹单元的层级信息，基于层级信息对当前顺序轨迹单元进行层级标记；按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；基于标注了层级信息的轨迹单元生成电子地图。采用本方法能够提升电子地图的准确性。

Description

电子地图生成方法、装置、计算机设备和存储介质 技术领域
本申请涉及一种电子地图生成方法、装置、计算机设备和存储介质。
背景技术
随着科技的发展，电子地图已经成为生活中必不可少的一种工具。现有的电子地图生成方法通常可分为数据采集，数据预处理，地图生成，语义信息提取，数据后处理等几个步骤。
在城市道路中，常常会出现如高架桥，地下停车场等多层级的行驶空间。
但是目前，针对多层级行驶空间采集得到的三维数据的预处理过程并不准确，从而影响电子地图的准确性。
发明内容
根据本申请公开的各种实施例，提供一种电子地图方法、装置、计算机设备和存储介质。
一种电子地图生成方法，其特征在于，包括：
获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
基于标注了层级信息的轨迹单元生成电子地图。
一种电子地图生成装置，其特征在于，所述装置包括：
路径轨迹获取模块，用于获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
层级标记模块，用于按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；根据当前顺序轨迹单元的空间 位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
电子地图生成模块，用于基于标注了层级信息的轨迹单元生成电子地图。
一种计算机设备，包括存储器和一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述一个或多个处理器执行以下步骤：
获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
基于标注了层级信息的轨迹单元生成电子地图。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
基于标注了层级信息的轨迹单元生成电子地图。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为一个实施例中电子地图生成方法的应用场景图；
图2为一个实施例中电子地图生成方法的流程示意图；
图3A为一个实施例中空间坐标系鸟瞰示意图；
图3B为一个实施例中空间坐标系三维示意图；
图4为一个实施例的基于网格地图生成电子地图的流程示意图；
图5为一个实施例中电子地图生成装置的框图；
图6为另一个实施例中电子地图生成装置的框图；
图7为一个实施例中计算机设备的框图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供的电子地图生成方法，可以应用于如图1所示的应用环境中。其中，测绘车102通过网络与服务器104进行通信。测绘车102是指安装有多种传感器，并基于多种传感器采集现场数据的车辆。测绘车102采集轨迹单元，并将采集得到的轨迹单元发送至服务器104，由服务器104对轨迹单元进行预处理，并基于预处理后的轨迹单元生成电子地图。其中，服务器104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在一个实施例中，如图2所示，提供了一种电子地图生成方法，以该方法应用于图1中的服务器为例进行说明，包括以下步骤：
S202，获取路径轨迹；路径轨迹包括至少一个轨迹单元；轨迹单元包括采集时间以及空间位置信息；路径轨迹由至少一个轨迹单元按照采集时间排序而得。
其中，轨迹单元是指安装于测绘车中的多种传感器采集得到的数据合集。轨迹单元包括空间位置信息、三维点云图像、采集空间位置信息和三维点云信息的采集时间。其中，空间位置信息是指测绘车在笛卡尔坐标系中的位置坐标；三维点云图像是指由架设在测绘车上的激光传感器采集得到的点云数据；采集时间是指测绘车采集得到空间位置信息以及三维点云图像的时间点。容易理解的，同一轨迹单元中的空间位置信息的采集时间和三维点云图像的采集时间一致。
具体地，可以预先在测绘车上架设激光传感器和GPS(Global Positioning System全球定位***)定位***，由激光传感器按照预设采集频率向行车区域发射探测信号，并将行车区域内物体反射回的信号与探测信号进行比对，得到周围的环境数据，基于环境数据 生成三维点云图像；以及由GPS定位***按照相同的预设采集频率确定当前位置的经纬度坐标和海拔高度，将当前位置的经纬度坐标和海拔高度转换为空间位置信息。
当测绘车接收到服务器发送的数据采集指令时，测绘车控制GPS定位***确定当前位置的经纬度坐标和海拔高度，并以架设在测绘车上的激光传感器的中心为原点建立空间坐标系，将当前位置的经纬度坐标和海拔高度转换为空间坐标系中的三维空间位置坐标。其中，参考图3，如图3所示，建立的空间坐标系以架设在测绘车上的激光传感器的中心为原点；以与地面水平的水平面为基准面(即以与地面水平的水平面为XOY平面)；以与测绘车运动方向水平的轴线为Y轴；以处于基准面中的、经过原点并与Y轴垂直的轴线为X轴；以经过原点并与基准面垂直的轴线为Z轴。图3A为一个实施例中空间坐标系鸟瞰示意图。图3B为一个实施例中空间坐标系三维示意图。容易理解的，在接收到服务器发送的数据采集停止指令之前，GPS定位***均以在接收到数据采集指令时建立空间坐标为基准，对经纬度坐标和海拔高度进行转换。
与此同时，当接收到服务器发送的数据采集指令时，测绘车控制激光传感器采集周围环境数据，并基于周围环境数据生成三维点云图像。当获取得到同一时刻采集得到的三维点云图像和空间位置信息时，测绘车综合三维点云图像、空间位置信息、三维点云图像和空间位置信息的采集时间，得到轨迹单元，并将在采集时段内采集得到的至少一个轨迹单元发送至服务器。
进一步地，服务器接收在采集时段内采集得到的至少一个轨迹单元，并基于轨迹单元中的采集时间对接收到的至少一个轨迹单元进行排序，得到路径轨迹。比如，当测绘车在采集时段内行经一个多层停车场时，测绘车按照预先设置的采集频率采集自身在多层停车场中的空间位置信息和多层停车场的三维点云图像，并将空间位置信息、三维点云图像和采集时间组合成轨迹单元。当采集结束后，测绘车按照采集时间升序进行对多个轨迹单元进行排序，得到在多层停车场中的路径轨迹。
在一个实施例中，由于GPS定位***与激光传感器的采集频率可能出现误差，因此，也可将采集时间差在预设差值范围内的空间定位信息以及三维点云图像组合为同一轨迹单元，并将空间定位信息的采集时间或三维点云图像的采集时间，作为轨迹单元中的采集时间。比如，当A空间定位信息的采集时间为a，B三维点云图像的采集时间为b，当a与b之间的时间差小于预设时间阈值时，组合A、B以及a，得到轨迹单元。
在一个实施例中，由于卫星信号遮挡，空间位置信息常常不准确，从而导致生成的电子地图产生重影。为了提高空间位置信息的精度，可采用直接法，光流法、卡尔曼滤波器，图优化等算对由GPS定位***采集得到的空间位置信息进行优化。
S204，按照排列顺序，确定路径轨迹中的当前顺序轨迹单元，以及位于当前顺序轨迹单元之前的至少一个前序轨迹单元。
其中，前序轨迹单元是指采集时间位于当前顺序轨迹单元的采集时间之前的数据。
具体地，当基于采集时间对轨迹单元进行排序，得到路径轨迹后，服务器对路径轨迹 中除位于首位的轨迹单元进行遍历，将当前遍历顺序的轨迹单元作为当前顺序轨迹单元，将路径轨迹中的位于当前顺序轨迹单元之前的数据作为前序轨迹单元，如此，直至最后顺序轨迹单元。例如，在开始对路径轨迹中的轨迹单元进行遍历时，服务器将路径轨迹中，位于第二顺序的轨迹单元作为当前顺序轨迹单元，将第一顺序的轨迹单元作为前序轨迹单元。又例如，当遍历至路径轨迹中的位于第三顺序的轨迹单元时，服务器将第三顺序的轨迹单元作为当前顺序轨迹单元，将第一顺序轨迹单元以及第二顺序轨迹单元均作为前序轨迹单元。
S206，根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
具体地，服务器提取当前顺序轨迹单元中的空间位置信息，以及提取每个前序轨迹单元中的空间位置信息，将当前顺序轨迹单元中的空间位置信息分别与每个前序轨迹单元中的空间位置信息进行信息匹配，当至少一个前序点数据中存在匹配成功的前序轨迹单元时，服务器将匹配成功的前序轨迹单元作为匹配轨迹单元。比如，将X轴之间的距离、Y轴之间的距离均小于预设距离阈值，以及Z轴之间的距离大于预设距离阈值的前序轨迹单元作为匹配轨迹单元。
在一个实施例中，空间位置信息包括水平面坐标和垂直坐标；根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元，包括：根据水平面坐标，确定当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；根据垂直坐标，确定当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；根据水平间距和垂直间距，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
其中，水平面是指空间坐标系中的XOY平面；X是指空间坐标系中的X轴；Y是指空间坐标系中的Y轴；O是指空间坐标系中的原点。水平面坐标是指测绘车在XOY平面中的位置坐标。垂直坐标是指空间坐标系中Z轴坐标。比如，当测绘车在空间坐标系中的空间位置坐标为(x，y，z)时，(x，y)即为水平面坐标；(z)即为垂直坐标。
具体地，服务器提取当前顺序轨迹单元中的水平面坐标，以及分别提取每个前序轨迹单元中的水平面坐标，并根据预设公式
Figure PCTCN2020105790-appb-000001
计算当前顺序轨迹单元与不同前序轨迹单元之间的水平间距。其中，x a是指当前顺序轨迹单元的水平面坐标中的X轴坐标值，即当前顺序轨迹单元中的空间信息中的X轴坐标值；x b是指前序轨迹单元的水平面坐标中的X轴坐标值，即前序轨迹单元中的空间信息中的X轴坐标值；y a当前顺序轨迹单元的水平面坐标中的Y轴坐标值，即当前顺序轨迹单元中的空间信息中的X轴坐标值同理，y b是指前序轨迹单元的水平面坐标中的Y轴坐标值，即前序轨迹单元中的空间信息中的Y轴坐标值。
例如，在当前顺序轨迹单元的水平面坐标为(x 1，y 1)，A前序轨迹单元的水平面坐标为(x 2，y 2)，B前序轨迹单元的水平面坐标为(x 3，y 3)时，当前顺序轨迹单元与A
Figure PCTCN2020105790-appb-000002
进一步地，服务器提取当前顺序轨迹单元中的垂直坐标，以及分别提取每个前序轨迹单元中的垂直坐标，并根据预设公式|z a-z b|计算当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距。其中，z a是指当前顺序轨迹单元的垂直坐标，即当前顺序轨迹单元中的空间信息中的Z轴坐标值；z b是指前序轨迹单元的垂直坐标值，即前序轨迹单元中的空间信息中的Z轴坐标值。
例如，在当前顺序轨迹单元的垂直坐标为(z 1)，A前序轨迹单元的垂直坐标为(z 2)，B前序轨迹单元的垂直坐标为(z 3)时，当前顺序轨迹单元与A前轨迹单元之间的垂直间距为|z 1-z 2|，当前顺序轨迹单元与B前序轨迹单元之间的垂直间距为|z 1-z 3|。
进一步地，服务器根据当前顺序轨迹单元与各前序轨迹单元之间的水平间距，以及当前顺序轨迹单元与各前序轨迹单元之间的垂直间距，从至少一个前序轨迹单元中筛选出匹配轨迹单元。
在一个实施例中，根据水平间距和垂直间距，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元，包括：将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
具体地，当获取得到当前顺序轨迹单元与各前序轨迹单元之间的水平间距，以及当前顺序轨迹单元与各前序轨迹单元之间的垂直间距时，服务器将水平间距小于预设水平阈值，以及垂直间距大于预设垂直阈值的前序轨迹单元作为匹配轨迹单元。在当前轨迹单元与前序轨迹单元之间的水平间距小于预设水平阈值，以及垂直间距大于预设垂直阈值时，表明测绘车在当前顺序轨迹单元的采集时刻与在匹配轨迹单元的采集时刻处于同一行驶空间，比如处于同一停车场中，或者处于同一高架桥中，并且测绘车在当前顺序轨迹单元的采集时刻所处高度与在匹配轨迹单元的采集时刻所处高度不一致，比如，测绘车在当前顺序轨迹单元的采集时刻处于高架桥上层，在匹配轨迹单元的采集时刻处于高架桥下层。
比如，在上述举例中，当预设水平阈值为r，预设垂直阈值为h，当前顺序轨迹单元的空间位置信息为(x 1,y 1,z 1)，A前序单元空间位置信息为(x 2,y 2,z 2)，B前序轨迹单元的空间位置信息为(x 3,y 3,z 3)，
Figure PCTCN2020105790-appb-000003
|z 1-z 2|＞h，|z 1-z 3|＜h时，服务器将A前序轨迹单元作为与当前顺序轨迹单元相匹配的匹配轨迹单元。
在一个实施例中，预设水平阈值以及预设垂直阈值可以根据需求自由设定，例如，可以根据行驶路径中的车道宽度设定预设水平阈值，根据行驶空间中的层高设置预设垂直阈值。
通过确定各轨迹单元的空间位置信息，可以基于空间位置信息确定与当前顺序轨迹单元处于同一行驶空间，但是高度值不一致的匹配轨迹单元，从而后续可以基于匹配轨迹单元的层级信息对当前顺序轨迹单元的层级信息进行标记。
S208，获取匹配轨迹单元的层级信息，基于层级信息对当前顺序轨迹单元进行层级标记。
其中，层级信息是指用以标识当前位置所在层级的信息，比如，可以基于层级信息确定当前测绘车所处楼层，也可以基于层级信息确定当前测绘车在高架桥上层或高架桥下层等；层级信息具体可以为层级标识。
具体地，当获取得到路径轨迹时，服务器对路径轨迹中的全部轨迹单元设置一初始层级信息，即服务器将路径轨迹中的所有轨迹单元的层级信息均设置为同一初始值，比如，服务器将路径轨迹中的所有轨迹单元的层级标识均设置为0。当获取得到匹配轨迹单元的层级信息时，服务器基于匹配轨迹单元的层级信息修改当前顺序轨迹单元的初始层级信息。
在一个实施例中，当基于空间位置信息确定，当前顺序轨迹单元的Z轴坐标值大于匹配轨迹单元的Z轴坐标值时，服务器增大匹配轨迹单元的层级信息，并将增大后的层级信息作为当前顺序轨迹单元的层级信息。比如，在上述举例中，当测绘车针对多层停车场进行数据采集，当|z 1-z 2|＞h，并且z 1-z 2＞0时，表明测绘车在当前轨迹单元时所处的行驶空间高于在匹配轨迹单元时所处的形式空间，即表明测绘车在当前轨迹单元时所处的停车场楼层高于在匹配轨迹单元时所处的停车场楼层，此时服务器获取匹配轨迹单元的层级标识，并对应增大获取得到的层级标识，将增大后的层级标识作为当前顺序轨迹路径的层级标识。
在当前顺序轨迹单元的Z轴坐标值小于匹配轨迹单元的Z轴坐标值时，服务器减小匹配轨迹单元的层级信息，并将减小后的层级信息作为当前顺序轨迹单元的层级信息。
S210，按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元。
具体地，当服务器确定当前顺序轨迹单元的层级信息后，服务器将路径轨迹中与当前顺序轨迹单元相邻，并且位于当前顺序轨迹单元之后的轨迹单元作为下一顺序轨迹单元，并将下一顺序轨迹单元作为当前顺序轨迹单元，返回确定位于当前顺序轨迹单元之前的至少一个前序轨迹单元的步骤，直至最后顺序轨迹单元。
S212，基于标注了层级信息的轨迹单元生成电子地图。
具体地，服务器按照层级信息对路径轨迹中的轨迹单元进行分组，将具有相同层级信息的轨迹单元归为一组，得到至少一组轨迹单元组。服务器获取与每组轨迹单元组相对应的网格地图模板，将轨迹单元组中的轨迹单元投影至对应的网格地图模板中，从而实现对网格地图模板中的各个网格图像进行渲染，得到网格地图。其中，网格地图是由M行N列的大小相同的网格图像组成，M与N均为正整数；网格地图的相关信息包括：网格地 图的行数和列数，网格地图的单位像素长度，网格地图中每个网格图像的横向像素和纵向像素，其中，网格地图的单位像素长度用于表征网格地图中一个像素所代表的物理长度，比如，网格地图的单位像素长度3厘米的含义就是在网格地图中每个像素所代表的物理长度是3厘米。网格地图模板是指网格图像未经渲染的网格地图，比如，网格地图模板可以是由多个空白网格图像组合而成的网格地图。
进一步地，服务器将渲染后的网格地图输入已训练完成的语义识别模型，通过语义识别模型提取网格地图中的语义信息。其中，语义信息是指网格地图中的交通信息。交通信息包括但不限于以下至少一项：道路形状信息、道路坡度信息、道路曲率信息、道路方向信息、车道宽度信息、防撞护栏信息、道路边缘信息、车道线信息、导流线信息等。
进一步地，服务器将语义信息加载至对应的网格地图中，得到每个层级轨迹单元组各自对应的电子地图。比如，当服务器生成与A轨迹单元组相对应的网格地图A，以及生成与B轨迹单元组相对应的网格地图B时，服务器基于语义识别模型提取出A网格地图中的语义信息A，以及提取出B网格地图中的语义信息B，并将语义信息A加载至A网格地图，得到与A轨迹单元组相对应的电子地图A，以及将语义信息B加载至B网格地图，得到与B轨迹单元组相对应的电子地图B。
在一个实施例中，服务器组合与各轨迹单元组相对应的电子地图，得到多层级空间下的电子地图。
在本实施例中，通过获取路径轨迹，可以按照路径轨迹中各轨迹单元的排列顺序确定当前顺序轨迹单元，以及前序轨迹单元；通过确定当前顺序轨迹单元以及前序轨迹单元，可以获取当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，以及根据所确定的空间位置信息从前序轨迹中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元；通过确定相匹配的匹配轨迹单元，可以根据匹配轨迹单元的层级信息对当前顺序轨迹单元进行层级标记，如此对路径轨迹中的轨迹单元进行迭代，从而实现对路径轨迹中的各轨迹单元的层级标记，进而可以根据标记了层级信息的轨迹单元生成对应的电子地图。由于本申请对路径轨迹中的轨迹单元均添加了层级信息，因此在生成电子地图的过程中，将针对多层级行驶空间采集得到的三维轨迹单元降维成为二维数据时，可以根据层级信息对降维后重叠的轨迹单元进行区分，从而减少了因轨迹单元重叠而导致生成的电子地图重叠的概率，提升了电子地图的准确性。
在一个实施例中，获取匹配轨迹单元的层级信息，基于层级信息对当前顺序轨迹单元进行层级标记，包括：当具有至少两个匹配轨迹单元时，根据匹配轨迹单元的空间位置信息从至少两个匹配轨迹单元中筛选出目标轨迹单元；获取目标轨迹单元的层级信息；根据目标轨迹单元的层级信息对当前顺序轨迹单元进行层级标记。
具体地，当获取得到匹配轨迹单元时，服务器统计获取得到的匹配轨迹单元的数量值。当具有两个或两个以上的与当前顺序轨迹单元相匹配的匹配轨迹单元时，服务器提取各匹配轨迹单元中的垂直坐标，并计算各垂直坐标的绝对值，将具有最大绝对值的匹配轨迹单 元作为目标轨迹单元。服务器获取目标轨迹单元的层级信息，并根据目标轨迹单元的层级信息对当前顺序轨迹单元进行层级标记。
本实施例中，通过从多个匹配轨迹单元中筛选出目标轨迹单元，使得服务器可以直接根据目标轨迹单元的层级信息调整当前顺序轨迹单元的层级信息，如此，提升了当前顺序轨迹单元的层级信息的确定效率。
在一个实施例中，空间位置信息包括垂直坐标；层级信息为层级标识；根据目标轨迹点的层级信息对当前顺序轨迹点进行层级标记，包括：当确定当前顺序轨迹单元的垂直坐标大于目标轨迹单元的垂直坐标时，增大目标轨迹单元的层级标识；基于增大后的层级标识对当前顺序轨迹单元进行层级标记。
具体地，服务器提取目标轨迹单元中的空间位置信息中的垂直坐标，以及提取出当前顺序轨迹单元中的空间位置信息中的垂直坐标，在当前顺序轨迹单元的垂直坐标大于目标轨迹单元的垂直坐标时，服务器获取目标轨迹单元的层级标识，以及预设的增加幅值，基于增加幅值增大目标轨迹单元的层级标识，将增大后的层级标识作为当前顺序轨迹单元的层级标识。比如，在目标轨迹单元的层级标识为l 1，增加幅值为l时，当前顺序轨迹单元的层级标识l 2即为l 1+l，目标轨迹单元的层级标识保持不变为l 1。
本实施例中，在当前顺序轨迹单元的垂直坐标大于目标轨迹单元的垂直坐标时，表明测绘车在当前顺序轨迹单元的采集时间下，所处的空间位置高于在目标轨迹单元的采集时间下的空间位置，此时，只需对应增大目标轨迹单元的层级标识，即可快速得到当前顺序轨迹单元的层级标识，大大提升了当前顺序轨迹单元的层级标识的确定效率。
在一个实施例中，上述电子地图生成方法还包括：当确定当前顺序轨迹单元的垂直坐标小于目标轨迹单元的垂直坐标时，降低目标轨迹单元的层级标识；将降低后的层级标识作为当前顺序轨迹单元的层级标记。
具体地，服务器提取目标轨迹单元中的空间位置信息中的垂直坐标，以及提取出当前顺序轨迹单元中的空间位置信息中的垂直坐标，在当前顺序轨迹单元的垂直坐标小于目标轨迹单元的垂直坐标时，表明测绘车在当前顺序轨迹单元的采集时间下，所处的空间位置低于在目标轨迹单元的采集时间下的空间位置。此时，服务器获取目标轨迹单元的层级标识，以及预设的降低幅值，基于降低幅值降低目标轨迹单元的层级标识，将降低后的层级标识作为当前顺序轨迹单元的层级标识。比如，在目标轨迹单元的层级标识为l 1，降低幅值为l时，当前顺序轨迹单元的层级标识l 2即为l 1-l，目标轨迹单元的层级标识保持不变为l 1。
本实施例中，在当前顺序轨迹单元的垂直坐标小于目标轨迹单元的垂直坐标时，只需对应降低目标轨迹单元的层级标识，即可快速得到当前顺序轨迹单元的层级标识，从而提升了当前顺序轨迹单元的层级标识的确定效率。
在一个实施例中，基于标注了层级信息的轨迹单元生成电子地图，包括：
S402，获取生成网格地图的参数信息。
S404，根据层级信息对路径轨迹中的轨迹单元进行分组，得到至少一个轨迹单元组。
S406，根据参数信息，生成每个轨迹单元组各自对应的网格地图。
S408，根据每个轨迹单元组各自对应的网格地图，生成电子地图。
其中，轨迹单元包括由激光传感器针对行车区域采集得到的三维点云图像。三维点云图像包括三维点数据的三维坐标和反射值。其中，三维点数据的三维坐标是指经激光扫描得到的行车区域中的反射点在空间坐标系中的坐标值。反射值包括颜色值和/或亮度值。例如，反射值可以是色彩模式中的三原色；反射值还可以是色彩模式中的色调值、饱和度值和明度值；反射值亦可以是色彩模式中的色相值、饱和度值和亮度值。
参数信息包括网格地图的行数和列数、网格地图的单位像素长度和网格地图中每个网格图像的横向像素和纵向像素。
具体地，服务器对路径轨迹中的各轨迹单元进行分组，将具有相同层级信息的轨迹单元作为一个轨迹单元组，以及获取预设的生成网格地图的参数信息。服务器对各轨迹单元组进行遍历，根据参数信息生成轨迹单元组各自对应的网格地图，综合每个轨迹单元组各自对应的网格地图，得到路径轨迹的网格地图。服务器将与路径轨迹相对应的网格地图输入预训练的语义识别模型中，基于语义识别模型生成路径轨迹的电子地图。其中，语义识别模型是指可以从网格地图中提取出语义信息的机器学习模型。
更具体地，服务器确定当前遍历顺序的轨迹单元组，以及生成网格地图的参数信息，根据参数信息生成与当前遍历顺序的轨迹单元组对应的网格地图模板。服务器对当前遍历顺序的轨迹单元组中的各轨迹单元进行遍历，将轨迹单元投影至网格地图模板中，得到与当前遍历顺序相对应的网格地图。
服务器确定当前遍历顺序的轨迹单元组中的当前遍历顺序的轨迹单元，并根据参数信息，以及当前遍历顺序轨迹单元中的三维点云图像中的各三维点数据的三维坐标，确定每个三维点数据在网格地图模板中的位置信息，比如，在当前三维点数据的三维坐标为(x 4，y 4，z 4)时，对应的网格地图模板中的位置信息为M 4行N 4列。服务器确定网格地图模板中与当前三维点数据具有相同位置信息的网格图像，并根据当前三维点数据中的反射值渲染具有相同位置信息的网格图像，从而实现将三维点数据投影至二维网格地图模板。服务器依次将三维点云图像中的三维点数据投影至网格地图模板，从而实现将当前遍历顺序轨迹单元投影至网格地图模板。
本实施例中，通过层级信息对路径轨迹中的轨迹单元进行分组，可以将具有相同层级信息的轨迹单元划分为同一组，从而可以将同一层级的轨迹单元投影至同一网格地图模板中，如此，可以减少将针对多层级行驶空间采集得到的轨迹单元投影至二维网格地图时，因高度信息的损失而导致轨迹单元的重叠的概率，进而提升了网格地图的准确性。
在一个实施例中，根据每个轨迹单元组各自对应的网格地图，生成电子地图，包括：分别将每个轨迹单元组各自对应的网格地图输入预训练的语义识别模型识，得到每个网格地图各自对应的语义信息；根据每个网格地图各自对应的语义信息，生成电子地图。
具体地，服务器预存储有已训练完成的语义识别模型。当生成每个轨迹单元组各自对应的网格地图时，服务器分别将网格地图输入语义识别模型，基于语义识别模型提取网格地图中的语义信息，并将语义信息加载至对应的网格地图中，得到每个轨迹单元组各自对应的电子地图。比如，当服务器基于语义识别模型从与A轨迹单元组对应的网格地图中提取出车道宽度信息，以及从与B轨迹单元组对应的网格地图中提取出道路边缘信息时，服务器将车道宽度信息加载至与A轨迹单元组对应的网格地图中，得到与A轨迹单元组对应的电子地图；将道路边缘信息加载至与B轨迹单元对应的网格地图中，得到与B轨迹单元对应的电子地图。进一步的，服务器综合各加载了语义信息的电子地图，得到路径轨迹的电子地图。
本实施例中，由于可以通过预训练的语义识别模型，准确地识别出网格地图中的语义信息，使得可以基于语义信息生成的电子地图更为精准。
在一个实施例中，上述电子地图生成方法包括：当在采集时段内采集得到多于一条路径轨迹时，确定多于一条路径轨迹中的各轨迹单元的采集时间；根据多于一条路径轨迹中的各轨迹单元的采集时间，确定多于一条路径轨迹各自对应的时间偏移量；按照时间偏移量，修改对应路径轨迹中的轨迹点的采集时间；根据修改后的采集时间，对多于一条路径轨迹进行合并，得到合并后的路径轨迹。
由于在对路径轨迹中的轨迹单元进行层级标记之前，需要按照采集时间对轨迹单元进行排序。当服务器接收到多辆测绘车在同一采集时间段内采集得到的多条路径轨迹，多条路径轨迹中的多个轨迹单元在采集时间上就会产生重叠，从而导致对多条路径轨迹进行合并，得到合并后的路径轨迹的空间位置会产生跳变。为了避免合并后的路径轨迹中的轨迹单元的采集时间不重叠，保证合并后的路径轨迹在时间和空间上都连续，可以对应修改的不同测绘车采集得到的轨迹单元的采集时间。
具体地，当在采集时段内接收到多辆测绘车发送的路径轨迹时，服务器对多条路径轨迹进性排序，得到轨迹序列。其中，对路径轨迹进行排序的方式可以依据需求自由设定，比如，根据接收到路径轨迹的时间点对路径轨迹进行排序。其中，多辆测绘车是指两辆或者两辆以上的测绘车；多条路径轨迹是指两条或者两条以上的路径轨迹。服务器按照排列顺序对轨迹序列中的路径轨迹进行遍历，确定当前遍历顺序路径轨迹，以及位于当前遍历顺序路径轨迹之前的前序路径轨迹。服务器确定每条前序路径轨迹中的轨迹单元的最大采集时间，根据每条前序路径轨迹中的轨迹单元的最大采集时间确定当前遍历顺序路径轨迹的时间偏移量，如此，服务器依次确定每条路径轨迹各自对应的时间偏移量。服务器基于时间偏移量修改对应路径轨迹的采集时间。
进一步地，服务器根据修改后的采集时间，对多条路径轨迹中的轨迹单元进行排序，从而对将多于一条路径轨迹进行合并，得到合并后的路径轨迹。
本实施例中，通过确定与各路径轨迹相对应的时间偏移量，并基于时间偏移量修改轨迹单元的采集时间，使得合并后的路径轨迹在时间上连续，从而后续可以基于时间连续的 路径轨迹生成对应的电子地图。
在一个实施例中，根据多于一条路径轨迹中的各轨迹点的采集时间，确定多于一条路径轨迹各自对应的时间偏移量，包括：根据多于一条路径轨迹中的轨迹单元的采集时间，对多于一条路径轨迹进行排序，得到轨迹序列；按照排列顺序对轨迹序列进行遍历；当轨迹序列中存在位于当前遍历顺序轨迹路径之前的至少一条前序轨迹路径时，获取每条前序轨迹路径中的最大采集时间；根据最大采集时间，确定当前遍历顺序的轨迹路径的时间偏移量。
具体地，当获取得到多于一条路径轨迹时，服务器获取每条路径轨迹中的最小采集时间，并根据最小采集时间，对多于一条路径轨迹进行排序，得到轨迹序列。比如，服务器按照最小采集时间升序对多于一条路径轨迹进行排序，得到轨迹序列。服务器按照排列顺序对轨迹序列中的路径轨迹进行遍历。服务器确定当前遍历顺序的路径轨迹，以及位于当前遍历顺序之前的全部前序路径轨迹，并将每条前序路径轨迹中的最大采集时间相加，得到当前遍历顺序的时间偏移量。当得到每条路径轨迹的时间偏移量后，服务器根据时间偏移量对应修改路径轨迹中的轨迹单元的采集时间。容易理解的，在当前遍历顺序路径轨迹不存在前序路径轨迹时，当前遍历顺序路径轨迹不存在时间偏移量。
比如，当服务器在0至10秒内接收到路径轨迹A、路径B以及路径轨迹C，A路径轨迹中的各轨迹单元的采集时间为[1,3,5,7]，B路径轨迹中的各轨迹单元的采集时间为[2,4,6,8]，C路径中的各轨迹单元的采集时间为[3,5,7,9]时，服务器按照各路径轨迹中的采集时间最小值对A路径轨迹、B路径轨迹以及C路径轨迹进行排序，得到轨迹序列。
服务器将A路径轨迹中的最大采集时间7作为路径轨迹的时间偏移量；将A路径轨迹中的最大采集时间7以及B路径轨迹中的最大采集时间8进行叠加，得到C路径轨迹的时间偏移量。服务器将B路径轨迹中的轨迹单元的采集时间均增加7，得到修改后的采集时间[9,11,13,15]；以及将C路径轨迹中的轨迹单元的采集时间均增加15，得到修改后的采集时间[18,20,22,24]。服务器按照修改后的采集时间对A路径轨迹、B路径轨迹以及C路径轨迹进行合并，从而合并后的路径轨迹中的各轨迹单元的采集时间为[1,3,5,7,9,11,13,15,18,20,22,24]，进而服务器基于上述电子地图生成方法对合并后的路径轨迹进行数据处理，得到与合并后的路径轨迹相对应的电子地图。
本实施例中，通过确定时间偏移量，使得后续可以基于时间偏移量修改路径轨迹的采集时间，进而可以基于修改后的采集时间对路径轨迹进行合并。
在一个实施例中，当服务器根据多于一条路径轨迹中的轨迹单元的采集时间，对多于一条路径轨迹进行排序，得到轨迹序列后，服务器按照排列顺序对轨迹序列进行遍历。服务器确定当前遍历顺序路径轨迹，以及与当前遍历顺序路径轨迹相邻，并且位于当前遍历顺序路径轨迹之前的一个前序路径轨迹，将前序路径轨迹中的最大采集时间作为当前遍历顺序路径轨迹的时间偏移量，并基于所确定的时间偏移量对当前遍历顺序路径轨迹中的各轨迹单元的采集时间进行修改。服务器根据排列顺序确定下一顺序路径轨迹，并将下一顺 序路径轨迹作为当前遍历顺序路径轨迹，返回确定与当前遍历顺序路径轨迹相邻，并且位于当前遍历顺序路径轨迹之前的一个前序路径轨迹的步骤，直至最后顺序路径轨迹。进一步地，服务器将采集时间修改后的路径轨迹进行合并，得到合并后的路径轨迹。
应该理解的是，虽然图2、4的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2、4中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在其中一个实施例中，如图5所示，提供了一种电子地图生成装置500，包括：路径轨迹获取模块502、层级标记模块504、地图生成模块506。
路径轨迹获取模块502，用于获取路径轨迹；路径轨迹包括至少一个轨迹单元；轨迹单元包括采集时间以及空间位置信息；路径轨迹由至少一个轨迹单元按照采集时间排序而得；
层级标记模块504，用于按照排列顺序，确定路径轨迹中的当前顺序轨迹单元，以及位于当前顺序轨迹单元之前的至少一个前序轨迹单元；根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元；获取匹配轨迹单元的层级信息，基于层级信息对当前顺序轨迹单元进行层级标记；按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
地图生成模块506，用于基于标注了层级信息的轨迹单元生成电子地图。
在一个实施例中，如图6所示，层级标记模块504还包括匹配轨迹单元确定模块5041，用于根据水平面坐标，确定当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；根据垂直坐标，确定当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；根据水平间距和垂直间距，从至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
在一个实施例中，匹配轨迹单元确定模块5041还用于将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
在一个实施例中，层级标记模块504还包括目标轨迹单元确定模块5042，用于当具有至少两个匹配轨迹单元时，根据匹配轨迹单元的空间位置信息从至少两个匹配轨迹单元中筛选出目标轨迹单元；获取目标轨迹单元的层级信息；根据目标轨迹单元的层级信息对当前顺序轨迹单元进行层级标记。
在一个实施例中，目标轨迹单元确定模块5042还用于当确定当前顺序轨迹单元的垂直坐标大于目标轨迹单元的垂直坐标时，增大目标轨迹单元的层级标识；将增大后的层级 标识作为当前顺序轨迹单元的层级标记。
在一个实施例中，目标轨迹单元确定模块5042还用于当确定当前顺序轨迹单元的垂直坐标小于目标轨迹单元的垂直坐标时，降低目标轨迹单元的层级标识；将降低后的层级标识作为当前顺序轨迹单元的层级标记。
在一个实施例中，地图生成模块506还包括网格地图生成模块5061，用于获取生成网格地图的参数信息；根据层级信息对路径轨迹中的轨迹单元进行分组，得到至少一个轨迹单元组；根据参数信息，生成每个轨迹单元组各自对应的网格地图；根据每个轨迹单元组各自对应的网格地图，生成电子地图。
在一个实施例中，网格地图生成模块5061还用于分别将每个轨迹单元组各自对应的网格地图输入预训练的语义识别模型识，得到每个网格地图各自对应的语义信息；根据每个网格地图各自对应的语义信息，生成电子地图。
在一个实施例中，电子地图生成装置500还包括合并模块508，用于当在采集时段内采集得到多于一条路径轨迹时，确定多于一条路径轨迹中的各轨迹单元的采集时间；根据多于一条路径轨迹中的各轨迹单元的采集时间，确定多于一条路径轨迹各自对应的时间偏移量；按照时间偏移量，修改对应路径轨迹中的轨迹点的采集时间；根据修改后的采集时间，对多于一条路径轨迹进行合并，得到合并后的路径轨迹。
在一个实施例中，合并模块508还用于根据多于一条路径轨迹中的轨迹单元的采集时间，对多于一条路径轨迹进行排序，得到轨迹序列；按照排列顺序对轨迹序列进行遍历；当轨迹序列中存在位于当前遍历顺序轨迹路径之前的至少一条前序轨迹路径时，获取每条前序轨迹路径中的最大采集时间；根据最大采集时间，确定当前遍历顺序的轨迹路径的时间偏移量。
在一个实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图7所示。该计算机设备包括通过***总线连接的处理器、存储器、网络接口和数据库。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作***、计算机可读指令和数据库。该内存储器为非易失性存储介质中的操作***和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储电子地图数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种电子地图生成方法。
本领域技术人员可以理解，图7中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
一种计算机设备，包括存储器和一个或多个处理器，存储器中储存有计算机可读指令，计算机可读指令被处理器执行时，使得一个或多个处理器执行时实现上述方法实施例中的步骤。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行时实现上述方法实施例中的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种电子地图生成方法，其特征在于，包括：
    获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
    按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
    根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
    获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
    按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
    基于标注了层级信息的轨迹单元生成电子地图。
  2. 根据权利要求1所述的方法，其特征在于，所述空间位置信息包括水平面坐标和垂直坐标；所述根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元，包括：
    根据所述水平面坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；
    根据所述垂直坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；
    根据水平间距和垂直间距，从所述至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
  3. 根据权利要求2所述的方法，其特征在于，所述根据水平间距和垂直间距，从所述至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元，包括：
    将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
  4. 根据权利要求1所述的方法，其特征在于，所述获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记，包括：
    当具有至少两个匹配轨迹单元时，根据匹配轨迹单元的空间位置信息从所述至少两个匹配轨迹单元中筛选出目标轨迹单元；
    获取所述目标轨迹单元的层级信息；
    根据所述目标轨迹单元的层级信息对所述当前顺序轨迹单元进行层级标记。
  5. 根据权利要求4所述的方法，其特征在于，所述空间位置信息包括垂直坐标；所述层级信息为层级标识；所述根据所述目标轨迹单元的层级信息对所述当前顺序轨迹点进行层级标记，包括：
    当确定所述当前顺序轨迹单元的垂直坐标大于所述目标轨迹单元的垂直坐标时，增大所述目标轨迹单元的层级标识；
    将增大后的层级标识作为当前顺序轨迹单元的层级标记。
  6. 根据权利要求5所述的方法，其特征在于，所述方法还包括：
    当确定所述当前顺序轨迹单元的垂直坐标小于所述目标轨迹单元的垂直坐标时，降低所述目标轨迹单元的层级标识；
    将降低后的层级标识作为当前顺序轨迹单元的层级标记。
  7. 根据权利要求1所述的方法，其特征在于，所述基于标注了层级信息的轨迹单元生成电子地图，包括：
    获取生成网格地图的参数信息；
    根据层级信息对所述路径轨迹中的轨迹单元进行分组，得到至少一个轨迹单元组；
    根据所述参数信息，生成每个轨迹单元组各自对应的网格地图；
    根据每个轨迹单元组各自对应的网格地图，生成电子地图。
  8. 根据权利要求7所述的方法，其特征在于，所述根据每个轨迹单元组各自对应的网格地图，生成电子地图，包括：
    分别将每个轨迹单元组各自对应的网格地图输入预训练的语义识别模型识，得到每个网格地图各自对应的语义信息；
    根据每个网格地图各自对应的语义信息，生成电子地图。
  9. 根据权利要求1-8任意一项所述的方法，其特征在于，所述方法还包括：
    当在采集时段内采集得到多于一条路径轨迹时，确定所述多于一条路径轨迹中的各轨迹单元的采集时间；
    根据所述多于一条路径轨迹中的各轨迹单元的采集时间，确定所述多于一条路径轨迹各自对应的时间偏移量；
    按照所述时间偏移量，修改对应路径轨迹中的轨迹点的采集时间；
    根据修改后的采集时间，对多于一条路径轨迹进行合并，得到合并后的路径轨迹。
  10. 根据权利要求9所述的方法，其特征在于，所述根据所述多于一条路径轨迹中的各轨迹点的采集时间，确定所述多于一条路径轨迹各自对应的时间偏移量，包括：
    根据所述多于一条路径轨迹中的轨迹单元的采集时间，对所述多于一条路径轨迹进行排序，得到轨迹序列；
    按照排列顺序对所述轨迹序列进行遍历；
    当所述轨迹序列中存在位于当前遍历顺序轨迹路径之前的至少一条前序轨迹路径时，获取每条前序轨迹路径中的最大采集时间；
    根据所述最大采集时间，确定当前遍历顺序的轨迹路径的时间偏移量。
  11. 一种电子地图生成装置，其特征在于，所述装置包括：
    路径轨迹获取模块，用于获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
    层级标记模块，用于按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
    地图生成模块，用于基于标注了层级信息的轨迹单元生成电子地图。
  12. 根据权利要求11所述的装置，其特征在于，所述空间位置信息包括水平面坐标和垂直坐标；所述层级标记模块包括匹配轨迹单元确定模块，用于根据所述水平面坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；根据所述垂直坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；根据水平间距和垂直间距，从所述至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
  13. 根据权利要求12所述的装置，其特征在于，所述匹配轨迹单元确定模块还用于将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
  14. 一种计算机设备，包括存储器及一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
    按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
    根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
    获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
    按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
    基于标注了层级信息的轨迹单元生成电子地图。
  15. 根据权利要求14所述的计算机设备，其特征在于，所述空间位置信息包括水平面坐标和垂直坐标；所述处理器执行所述计算机可读指令时还执行以下步骤：
    根据所述水平面坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；
    根据所述垂直坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；
    根据水平间距和垂直间距，从所述至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
  16. 根据权利要求15所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
  17. 根据权利要求14所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    当具有至少两个匹配轨迹单元时，根据匹配轨迹单元的空间位置信息从所述至少两个匹配轨迹单元中筛选出目标轨迹单元；
    获取所述目标轨迹单元的层级信息；
    根据所述目标轨迹单元的层级信息对所述当前顺序轨迹单元进行层级标记。
  18. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取路径轨迹；所述路径轨迹包括至少一个轨迹单元；所述轨迹单元包括采集时间以及空间位置信息；所述路径轨迹由至少一个轨迹单元按照采集时间排序而得；
    按照排列顺序，确定所述路径轨迹中的当前顺序轨迹单元，以及位于所述当前顺序轨迹单元之前的至少一个前序轨迹单元；
    根据当前顺序轨迹单元的空间位置信息和前序轨迹单元的空间位置信息，从所述至少一个前序轨迹单元中筛选出与所述当前顺序轨迹单元相匹配的匹配轨迹单元；
    获取所述匹配轨迹单元的层级信息，基于所述层级信息对当前顺序轨迹单元进行层级标记；
    按照排列顺序，确定下一顺序轨迹单元，将下一顺序轨迹单元作为当前顺序轨迹单元进行迭代，直至最后顺序轨迹单元；
    基于标注了层级信息的轨迹单元生成电子地图。
  19. 根据权利要求18所述的存储介质，其特征在于，所述空间位置信息包括水平面坐标和垂直坐标；所述计算机可读指令被所述处理器执行时还执行以下步骤：
    根据所述水平面坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的水平间距；
    根据所述垂直坐标，确定所述当前顺序轨迹单元与不同前序轨迹单元之间的垂直间距；
    根据水平间距和垂直间距，从所述至少一个前序轨迹单元中筛选出与当前顺序轨迹单元相匹配的匹配轨迹单元。
  20. 根据权利要求19所述的存储介质，其特征在于，所述计算机可读指令被所述处 理器执行时还执行以下步骤：
    将水平间距与垂直间距均符合预设条件的前序轨迹单元作为匹配轨迹单元。
PCT/CN2020/105790 2020-07-30 2020-07-30 电子地图生成方法、装置、计算机设备和存储介质 WO2022021209A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/105790 WO2022021209A1 (zh) 2020-07-30 2020-07-30 电子地图生成方法、装置、计算机设备和存储介质
CN202080099129.6A CN115668333A (zh) 2020-07-30 2020-07-30 电子地图生成方法、装置、计算机设备和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/105790 WO2022021209A1 (zh) 2020-07-30 2020-07-30 电子地图生成方法、装置、计算机设备和存储介质

Publications (2)

Publication Number Publication Date
WO2022021209A1 WO2022021209A1 (zh) 2022-02-03
WO2022021209A9 true WO2022021209A9 (zh) 2022-11-03

Family

ID=80036950

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/105790 WO2022021209A1 (zh) 2020-07-30 2020-07-30 电子地图生成方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN115668333A (zh)
WO (1) WO2022021209A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117789444A (zh) * 2022-09-19 2024-03-29 北京初速度科技有限公司 一种停车场数据的匹配方法、装置、设备、介质及车辆

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102313556B (zh) * 2010-07-01 2014-04-02 北京四维图新科技股份有限公司 环岛上的路径匹配的方法及装置
CN107545833B (zh) * 2016-06-24 2019-09-24 高德信息技术有限公司 一种道路绘制方法及装置
US20180052593A1 (en) * 2016-08-18 2018-02-22 Mapbox, Inc. Providing visual selection of map data for a digital map
CN106643779B (zh) * 2016-09-29 2019-07-19 合肥工业大学 一种基于改进的连通图遍历算法的路径图生成方法
CN111380543B (zh) * 2018-12-29 2023-05-05 沈阳美行科技股份有限公司 地图数据生成方法及装置

Also Published As

Publication number Publication date
WO2022021209A1 (zh) 2022-02-03
CN115668333A (zh) 2023-01-31

Similar Documents

Publication Publication Date Title
CN109059954B (zh) 支持高精度地图车道线实时融合更新的方法和***
JP7485749B2 (ja) ビデオベースの位置決め及びマッピングの方法及びシステム
US11085775B2 (en) Methods and systems for generating and using localisation reference data
CN106570446B (zh) 车道线提取的方法和装置
KR102143108B1 (ko) 차선 인식 모델링 방법, 장치, 저장 매체 및 기기, 및 인식 방법, 장치, 저장 매체 및 기기
WO2018068653A1 (zh) 点云数据处理方法、装置及存储介质
JP6595182B2 (ja) マッピング、位置特定、及び姿勢補正のためのシステム及び方法
CN102208013B (zh) 风景匹配参考数据生成***和位置测量***
CN111144388A (zh) 一种基于单目影像的道路标志标线的更新方法
JP5388082B2 (ja) 静止物地図生成装置
CN112346463B (zh) 一种基于速度采样的无人车路径规划方法
CN110197173B (zh) 一种基于双目视觉的路沿检测方法
WO2011160672A1 (en) Method for obtaining drivable road area
JP2012208525A (ja) 静止物地図生成装置
Liu et al. Image-translation-based road marking extraction from mobile laser point clouds
CN115344655A (zh) 地物要素的变化发现方法、装置及存储介质
US11828620B2 (en) Method of predicting road attributes, data processing system and computer executable code
WO2022021209A9 (zh) 电子地图生成方法、装置、计算机设备和存储介质
CN112833891A (zh) 基于卫片识别的道路数据与车道级地图数据的融合方法
CN110174115B (zh) 一种基于感知数据自动生成高精度定位地图的方法及装置
CN111754388B (zh) 一种建图方法及车载终端
CN113227713A (zh) 生成用于定位的环境模型的方法和***
Lee et al. Semi-automatic framework for traffic landmark annotation
CN113762195A (zh) 一种基于路侧rsu的点云语义分割与理解方法
Zou et al. Inertia mutation energy model to extract roads by crowdsourcing trajectories

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: 20946911

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 27/06/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20946911

Country of ref document: EP

Kind code of ref document: A1