CN116972859A - Global positioning method, device, equipment and storage medium based on contour map - Google Patents
Global positioning method, device, equipment and storage medium based on contour map Download PDFInfo
- Publication number
- CN116972859A CN116972859A CN202310750255.9A CN202310750255A CN116972859A CN 116972859 A CN116972859 A CN 116972859A CN 202310750255 A CN202310750255 A CN 202310750255A CN 116972859 A CN116972859 A CN 116972859A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- map
- cloud data
- value
- global positioning
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 96
- 239000011159 matrix material Substances 0.000 claims description 25
- 230000009466 transformation Effects 0.000 claims description 25
- 238000005457 optimization Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 238000009434 installation Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000013138 pruning Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 abstract description 17
- 230000006835 compression Effects 0.000 abstract description 3
- 238000007906 compression Methods 0.000 abstract description 3
- 238000004891 communication Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000005266 casting Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
- G01C21/30—Map- or contour-matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
- G01C21/3833—Creation or updating of map data characterised by the source of data
- G01C21/3841—Data obtained from two or more sources, e.g. probe vehicles
-
- 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
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
-
- 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
- G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
-
- 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
- G01S19/48—Determining position by combining or switching between position solutions derived from the satellite radio beacon positioning system and position solutions derived from a further system
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Navigation (AREA)
Abstract
The invention provides a global positioning method, a global positioning device, global positioning equipment and a storage medium based on a contour map, wherein the global positioning method comprises the following steps: acquiring first point cloud data and GNSS data acquired by a sensor; removing the point cloud ground points in the first point cloud data to obtain second point cloud data; loading a contour map grid-connected grid of the region of interest by using GNSS data to obtain an occupied grid map; searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning; and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result. The invention adopts the form of the contour map to express the point cloud data, realizes the compression of the point cloud map, saves the storage space and optimizes the positioning result by the maximum occupation probability. Therefore, the global high-precision positioning can be completed without a large amount of storage space and corresponding calculation force.
Description
Technical Field
The invention relates to the technical field of automatic driving of vehicles, in particular to a global positioning method, a global positioning device, global positioning equipment and a storage medium based on a contour map.
Background
In the industries of automatic driving, robots and the like, accurate global positioning is important to modules such as perception control and the like. High-precision Inertial Navigation (INS) cannot provide long-term reliable global positioning in multi-path multi-occlusion environments. 3D light detection and LiDAR ranging (LiDAR) systems are becoming smaller, cheaper and lighter, providing rich and accurate remote 3D information that can be used for global positioning in conjunction with pre-established high-precision a priori point cloud maps. However, a complete a priori point cloud map occupies a large amount of memory and has a large amount of useless data. Registration of feature-based maps is a difficult task due to different viewpoint awareness and occlusion, and efficient features are missing in high-speed driving environments.
Disclosure of Invention
The global positioning method, the device, the equipment and the storage medium based on the contour map are used for solving the problems that the precision of a positioning result is low and the calculation efficiency is low when a high-precision prior point cloud map is used in the prior art, and the global high-precision positioning can be completed by adopting a laser radar positioning strategy based on the contour map without depending on high-precision inertial navigation and without needing a large amount of storage space and corresponding calculation force.
The invention provides a global positioning method based on a contour map, which comprises the following steps:
acquiring first point cloud data and GNSS data acquired by a sensor;
removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
loading a contour map of the region of interest by using the GNSS data, and gridding the contour map to obtain an occupied grid map;
searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to obtain a final global positioning result.
According to the global positioning method based on the contour map provided by the invention, before the step of removing the point cloud ground points in the first point cloud data, the method further comprises the following steps: the downsampling process for the first point cloud data specifically includes:
multiplying the first point cloud data with a pre-calibrated sensor external parameter to obtain third point cloud data under a vehicle body coordinate system;
dividing the space region covered by the third point cloud data into a plurality of voxel grids, and dividing the points of the third point cloud data into corresponding voxel grids;
And representing all the point cloud points in the voxel grids through the centroid of each voxel grid so as to construct sparse point cloud data.
According to the global positioning method based on the contour map provided by the invention, the step of removing the point cloud ground points in the first point cloud data comprises the following steps:
ordering the point cloud points of the first point cloud data according to the Z-axis numerical value;
removing point cloud points with the Z-axis numerical value larger than the installation height of the sensor;
selecting a preset first number of point cloud points in sequence from small to large based on the Z-axis value and calculating a mean value;
traversing all the remaining point cloud points, calculating the difference value between the Z-axis value of the point cloud points and the mean value, sorting the point cloud points with the difference value smaller than a preset threshold value from small to large, and selecting a preset second number of point cloud points from the point cloud points as initial points of the point cloud ground;
and carrying out plane fitting on the initial points, and removing the point cloud points on the plane through a fitted plane equation to finish the removal of the point cloud ground points.
According to the global positioning method based on the contour map provided by the invention, the step of loading the contour map of the region of interest by using the GNSS data and gridding the contour map to obtain the occupied grid map comprises the following steps:
Acquiring an initial value of the current positioning of the vehicle based on the GNSS data, and determining a contour map according to the initial value;
rasterizing the contour map, and initializing the value of each grid unit in the grid map;
determining all grid points passed by the starting point position and the end point position of the contour line by using a light projection method, and adding the values of the grid units according to the current height of the contour line for each passed grid point by using a preset value until all contour lines in the contour line map are traversed;
and carrying out normalization processing on the value of each grid unit in the grid map to obtain a normalized grid map serving as the occupied grid map.
According to the global positioning method based on the contour map provided by the invention, the step of searching the vehicle pose in the preset range based on the second point cloud data and the occupied grid map to obtain the initial pose value of global positioning comprises the following steps:
creating a search tree for calculating the pose of the vehicle by adopting a branch-and-bound method, recursively searching a transformation matrix corresponding to each node on the search tree, and multiplying the transformation matrix by the second point cloud data to obtain third point cloud data;
Calculating grid coordinates corresponding to the point cloud points for each point cloud point in the third point cloud data, judging whether the grid coordinates are in the range of the occupied grid map, if so, taking the value of the grid unit of the occupied grid map as the score of the point cloud point, and if not, returning a negative value as the score of the point cloud point;
adding the scores of all the point cloud points to obtain a score term, calculating whether the score term is an optimal score, if so, taking the transformation matrix obtained by current search as an optimal transformation matrix, and if not, pruning nodes lower than the score term on the search tree;
and converting the second point cloud data into a world coordinate system according to the optimal transformation matrix so as to determine an initial pose value of global positioning of the vehicle.
According to the global positioning method based on the contour map provided by the invention, the step of obtaining the pose with the largest probability of occupying the occupied grid map by the second point cloud data to optimize the initial pose value comprises the following steps:
using a CERES optimizer, constructing in the CERES optimizer a maximum occupancy probability of the second point cloud data to the occupancy grid map;
And based on the maximum occupancy probability, registering the second point cloud data and the contour map by a nonlinear optimization method so as to optimize the initial pose value.
According to the global positioning method based on the contour map provided by the invention, the step of constructing the maximum occupation probability from the second point cloud data to the occupation grid map in the CERES optimizer by using the CERES optimizer comprises the following steps:
constructing a CERES optimizer, and importing the second point cloud data and the occupied grid map into the CERES optimizer;
and defining an interpolator in the CERES optimizer, interpolating each grid unit in the occupied grid map through the interpolator, and calculating the matching degree of the interpolated occupied grid map and the second point cloud data so as to determine the maximum occupation probability from the second point cloud data to the occupied grid map.
The invention also provides a global positioning device based on the contour map, which comprises:
the acquisition module is used for acquiring the first point cloud data and the GNSS data acquired by the sensor;
the ground point removing module is used for removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
The map gridding module is used for loading a contour map of the region of interest by using the GNSS data, and gridding the contour map to obtain an occupied grid map;
the initial pose value acquisition module is used for searching the pose of the vehicle in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and the initial pose value optimization module is used for acquiring the pose with the largest probability of occupying the grid map by the second point cloud data to optimize the initial pose value and taking the pose as a final global positioning result.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the global positioning method based on the contour map when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the global positioning method based on a contour map as set forth in any one of the above.
The invention provides a global positioning method, a global positioning device, global positioning equipment and a storage medium based on a contour map, wherein first point cloud data and GNSS data acquired by a sensor are acquired; removing the point cloud ground points in the first point cloud data to obtain second point cloud data; loading a contour map of the region of interest by using GNSS data, and meshing the contour map to obtain an occupied grid map; searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning; and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result. In the prior art, the problem that the positioning result is low in precision and low in calculation efficiency exists when the high-precision priori point cloud map is used for realizing positioning is solved, the point cloud data are represented in the form of the contour map, the compression of the point cloud map is realized, the storage space is saved, and a large amount of calculation space can be saved compared with the direct storage of the complete high-precision point cloud map; and for the rasterized contour map, the initial pose value of the vehicle is optimized through the maximum occupation probability, so that the positioning result is optimized. Therefore, the global high-precision positioning can be completed without a large amount of storage space and corresponding calculation force.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a global positioning method based on a contour map according to the present invention;
FIG. 2 is a second flow chart of a global positioning method based on a contour map according to the present invention;
FIG. 3 is a schematic diagram of a global positioning device based on a contour map according to the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Reference numerals:
31: an acquisition module; 32: a ground point removal module; 33: a map gridding module; 34: an initial pose value acquisition module; 35: and an initial pose value optimization module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be clearly and completely described in the following description with reference to specific embodiments of the present invention and the accompanying drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that those skilled in the art explicitly and implicitly understand that the described embodiments of the invention can be combined with other embodiments without conflict. Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs. The terms "a," "an," "the," and similar referents in the context of the invention are not to be construed as limiting the quantity, but rather as singular or plural. The terms "comprising," "including," "having," and any variations thereof, are intended to cover a non-exclusive inclusion; the terms "first," "second," "third," and the like, as used herein, are merely distinguishing between similar objects and not representing a particular ordering of objects.
The unmanned automobile can be provided with a laser radar, a network and driving control equipment.
The lidar may be a radar system that detects a characteristic amount of a position, a speed, or the like of a target with a laser beam emitted. Specifically, when a laser beam emitted from a laser radar irradiates a target surface, the reflected laser beam carries information of azimuth, distance, and the like. When the laser beam emitted by the laser radar is scanned according to a certain track, reflected laser point information is recorded while scanning, and a large number of laser points can be obtained due to extremely fine scanning, so that a point cloud can be formed.
The driving control equipment, also called an onboard brain, is responsible for intelligent control of the unmanned automobile. The driving control device can be a separately arranged controller, such as a programmable logic controller, a singlechip, an industrial controller and the like; the device can also be equipment consisting of other electronic devices with input/output ports and operation control functions; but also a computer device equipped with a vehicle driving control type application. The global positioning method based on the contour map provided by the embodiment of the invention is generally executed by the driving control equipment, and correspondingly, the global positioning device based on the contour map is generally arranged in the driving control equipment.
A network is provided between the lidar and the drive control device to provide a medium for the communication link. The network may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others. The lidar may interact with the driving control device via a network to receive or send messages or the like.
In the prior art, the map-based positioning schemes also adopt complete point cloud map schemes, and the positioning is performed by extracting vectorization features (such as lane lines, signboards and the like), and the problems of low calculation efficiency and the like caused by using the complete point cloud map are solved. Therefore, the invention provides a global positioning method which saves the calculation cost and is applicable to different scenes and based on the contour map.
Example 1
Referring to fig. 1, the global positioning method based on the contour map provided in this embodiment includes:
step S1: acquiring first point cloud data and GNSS data acquired by a sensor;
in this embodiment, the sensors may be a camera, a millimeter wave radar, and a laser radar, and the laser radar is mounted on the vehicle, for example, and the point cloud data may be obtained by scanning the environment in front of the vehicle in the traveling direction by the laser radar. The vehicle may be an autonomous vehicle or a assisted driving vehicle. Point clouds are sets of points in three-dimensional space; consisting of N D-dimensional points, which when d=3 can be represented as three-dimensional coordinate points (x, y, z), each of which is determined by a certain (xyz) position, we can assign other properties (e.g. RGB colors) to them at the same time. They are the original form of lidar data as it is acquired, and the stereo vision system and RGB-D data (containing images marked with depth values for each pixel point) are typically converted to a point cloud before further processing.
GNSS (global navigation satellite system) includes navigation satellite systems such as GPS, beidou, gnonass, galileo, and the like. Taking a GPS navigation satellite system as an example, GPS (Global Positioning System ) observation data refer to observation data acquired by a GPS signal receiver, and the GPS observation data provides rough position information and track information, so that some basic requirements of map making are met, and the method is applied to the field of automobile driving. It should be noted that the GPS observation data may be observation data obtained by a low-cost GPS receiver, for example, a low-cost GPS receiver with standard accuracy, or a high-accuracy GPS receiver. In the invention, the lane-level rapid positioning of the vehicle can be improved by using the observation data acquired by the low-cost GPS, and the cost of vehicle positioning can be reduced.
Step S2: removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
in this embodiment, step S2 specifically includes:
step S21: ordering the point cloud points of the first point cloud data according to the Z-axis numerical value;
step S22: removing point cloud points with the Z-axis numerical value larger than the installation height of the sensor;
step S23: selecting a preset first number of point cloud points in sequence from small to large based on the Z-axis value and calculating a mean value;
step S24: traversing all the remaining point cloud points, calculating the difference value between the Z-axis value and the mean value of the point cloud points, sorting the point cloud points with the difference value smaller than a preset threshold value from small to large, and selecting a preset second number of point cloud points as initial points of the point cloud ground;
step S25: and carrying out plane fitting on the initial points, and removing the point cloud points on the plane through a fitted plane equation to finish the removal of the point cloud ground points.
Specifically, the preset first number is set to 100 point cloud points, the preset second number is set to 50 point cloud points, and the preset threshold value is 0.1m; removing point cloud points larger than the installation height of the sensor, selecting 100 point cloud points according to the descending order based on the Z-axis numerical value, calculating the mean value, traversing all the rest point cloud points, calculating the difference value between the Z-axis numerical value and the mean value, if the difference value is within 0.1m, arranging the point cloud points according to the descending order of the difference value, selecting 50 point cloud points as initial points, performing plane fitting by using the 50 point cloud points, and removing the point cloud points on the plane, namely the point cloud ground point, by using a plane fitting equation so as to realize the removal of the point cloud ground point. Here, the estimated point cloud ground points are sampled, selected and then fitted, so that the calculated amount of plane fitting is effectively reduced. The first number and the second number, and the preset threshold may also be determined based on the calculation requirement, which is not limited herein.
In this embodiment, before step S2, the method further includes: the downsampling process for the first point cloud data specifically includes:
multiplying the first point cloud data with a pre-calibrated sensor external parameter to obtain third point cloud data under a vehicle body coordinate system;
dividing a space region covered by the third point cloud data into a plurality of voxel grids, and dividing points of the third point cloud data into corresponding voxel grids;
all the point cloud points in the voxel grid are represented by the centroid of each voxel grid to construct sparse point cloud data.
Specifically, the original point cloud data acquired by the sensor contains partial noise and outliers, and has the problems of irregular point cloud density and the like, so voxel filtering is required to be performed on the first point cloud data, a three-dimensional voxel grid is created for the input point cloud by voxel filtering, the centroid of all points in a voxel in each voxel grid is used for approximately representing other point cloud points in the voxel grid, and therefore all points in the voxel grid are finally represented by a centroid point. In this way, the amount of processed data can be reduced, the shape of the original point cloud can be maintained in the down-sampling process, and interference caused by noise point cloud points can be avoided in the following process of filtering ground points.
Step S3: loading a contour map of the region of interest by using GNSS data, and meshing the contour map to obtain an occupied grid map;
in this embodiment, loading a contour map of a region of interest with GNSS data, modeling and rasterizing the contour map by using a ray casting method to obtain an occupied grid map, including:
step S31: acquiring an initial value of the current positioning of the vehicle based on the GNSS data, and determining a contour map according to the initial value;
step S32: rasterizing the peer-to-peer high-line map, and initializing the value of each grid unit in the grid map;
step S33: determining all grid points passing by the starting point position and the end point position of the contour line by using a light ray projection method, and adding the values of the grid units according to the height of the current contour line for each passing grid point by using a preset value until all contour lines in the contour line map are traversed;
step S34: and carrying out normalization processing on the value of each grid unit in the grid map to obtain a normalized grid map serving as an occupied grid map.
Specifically, an initial position of the current vehicle positioning is obtained according to GNSS data, a circle with the initial position as a circle center and a radius of 100m is used as a circle, and a corresponding contour map is obtained from a high-precision map engine. The grids are used for rasterizing the peer-to-peer altitude map according to the square with the size of 5cm, and the value of each grid unit in the grid map is initialized to be 0. And determining all grid points passed by the starting point position and the end point position of a single contour line in the contour line map by using a light projection method, judging the height of the current contour line every time the grid points pass, if the height of the contour line is greater than or equal to 2m, adding 1 to the value of the grid unit where the contour line is positioned, otherwise adding 0.5 to the value of the grid unit, and calculating. And carrying out normalization processing on the value of each grid unit in the grid map to obtain a normalized grid map serving as an occupied grid map. And (3) carrying out re-modeling on the contour map by adopting a Ray casting (Ray-casting) method, and realizing the effective representation of the contour map in the follow-up global positioning algorithm optimization.
Step S4: searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
in this embodiment, a branch-and-bound method is adopted to search the pose of the vehicle to obtain an initial pose value for global positioning, and correspondingly, step S4 includes:
step S41: creating a search tree for calculating the pose of the vehicle by adopting a branch-and-bound method, recursively searching a transformation matrix corresponding to each node on the search tree, and multiplying the transformation matrix by the second point cloud data to obtain third point cloud data;
step S42: for each point cloud point in the third point cloud data, calculating grid coordinates corresponding to the point cloud point, judging whether the grid coordinates are in the range of occupying the grid map, if so, taking the value of a grid unit occupying the grid map as the score of the point cloud point, and if not, returning a negative value as the score of the point cloud point;
step S43: accumulating the scores of all the points and points to obtain a score term, and calculating whether the score term is an optimal score, if so, taking a transformation matrix obtained by current search as an optimal transformation matrix, and if not, pruning nodes lower than the score term on a search tree;
Step S44: and converting the second point cloud data into a world coordinate system according to the optimal transformation matrix to determine an initial pose value of global positioning of the vehicle.
In particular, because the range is too large, the search space may also be too large, resulting in slower computations. Therefore, the search range is determined based on the confidence of the prior position, the position (x, y) and the heading angle (yaw) of the vehicle within a certain range are searched, for example, the search range in the x direction is + -10 m, the search range in the y axis is + -4 m, the search range in the yaw angle is + -0.1 rad, the search range is narrowed, and the calculation amount of the algorithm is further reduced.
The principle of the branch-and-bound algorithm is to represent all possible solutions of a problem as nodes in a tree, where the root node is represented as all possible solutions, the child nodes of each node constitute a subset of the parent nodes, and the leaf nodes are of a single structure, representing one possible solution of the problem. In the invention, a plurality of leaf nodes are rapidly evaluated through depth-first search, and the nodes with the score lower than the optimal score are pruned to rapidly find the optimal match. The method comprises the steps of initializing a global positioning and optimizing problem of a vehicle by using a branch and bound method, searching an initial pose value of global positioning of the vehicle on a rasterized contour map, taking the normalized contour map as input of a branch and bound algorithm, and multiplying the contour map with second point cloud data based on a transformation matrix when the algorithm starts searching to obtain third point cloud data; and for each point cloud point in the third point cloud data, calculating grid coordinates corresponding to the point cloud point, judging whether the grid coordinates are in the range of occupying the grid map, if so, taking the value of a grid unit occupying the grid map as the score of the point cloud point, otherwise, returning a negative value as the score of the point cloud point, and accumulating the scores of all the point cloud points to obtain a score. And recursively searching a transformation matrix corresponding to each node by using a branch-and-bound method, and taking the leaf node with the optimal score as an optimal transformation matrix, thereby multiplying second point cloud data observed by a sensor according to the optimal transformation matrix, and calculating an initial pose value of global positioning of the vehicle. Compared with the iterative closest method (ICP), the method only can perform small-range optimization under the condition of a given initial value, and when the initial value error is large, the optimal solution cannot be calculated, and the optimal real position is given.
Step S5: and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result.
In this embodiment, step S5 specifically includes:
step S51: using a CERES optimizer, constructing second point cloud data in the CERES optimizer to occupy a maximum occupancy probability of the grid map;
in this embodiment, step S51 specifically includes:
step S511: constructing a CERES optimizer, and importing the second point cloud data and the occupation grid map into the CERES optimizer;
step S512: and defining an interpolator in the CERES optimizer, interpolating each grid unit in the occupied grid map through the interpolator, and calculating the matching degree of the interpolated occupied grid map and the second point cloud data so as to determine the maximum occupied probability of the second point cloud data to the occupied grid map.
Step S52: and based on the maximum occupancy probability, registering the second point cloud data and the contour map by a nonlinear optimization method so as to optimize the initial pose value.
Specifically, a CERES optimizer is constructed, and the second point cloud data and the occupancy grid map are imported into the CERES optimizer. Defining an interpolator (BiCubicInterpolator) in the CERES optimizer, carrying out map bilinear interpolation on each grid unit in the occupied grid map through the interpolator, and calculating the matching degree of the interpolated occupied grid map and the second point cloud data (the maximum matching degree of the point cloud data and the contour map is the maximum occupancy probability), namely acquiring an evaluation value by using the BiCubicInterpolator in the process of evaluating the matching probability by using the evaluation function, and subtracting the acquired evaluation value by using 1 (the maximum matching degree is 1) to obtain a cost value which is taken as the maximum occupancy probability of the second point cloud data to the occupied grid map. And registering the second point cloud data and the contour map by a nonlinear optimization method (LM, levenberg-Marquardt) to optimize the initial pose value. According to the invention, the modeled contour map is subjected to bilinear interpolation, and the CERES optimizer is used for further fine optimization of the existing global positioning optimization problem, so that high-precision positioning in the contour map is realized.
In summary, in order to solve the problems of low accuracy of positioning results and low calculation efficiency caused by using a high-accuracy priori point cloud map to realize positioning in the prior art, the global positioning method based on the contour map provided by the embodiment re-represents the point cloud map in a contour form, so that the compression of the point cloud map is realized, the storage space is saved, and a large amount of calculation space can be saved compared with the direct storage of the complete high-accuracy point cloud map; and initializing a positioning problem by using branch delimitation for the rasterized contour map, realizing the rapid initialization of the positioning problem, carrying out bilinear interpolation on the modeled occupied grid map, and obtaining the maximum occupancy probability for the initial pose value in a CERES optimizer by a nonlinear optimization method so as to optimize the initial pose value of the vehicle and realize the optimization of the positioning result. It is less likely to fall into locally optimal solutions and more computation performance efficient than using iterative nearest neighbor (ICP). Therefore, the global high-precision positioning can be completed without a large amount of storage space and corresponding calculation force.
Referring to fig. 2, fig. 2 is a second flow chart of the global positioning method based on the contour map according to the present invention. The global positioning method based on the contour map provided by the embodiment of the invention comprises the following steps:
Step A1: GPS data acquired by a low-cost GPS;
step A2: loading a contour map of the region of interest;
step A3: obtaining an occupied grid map by adopting a ray projection method for the peer-to-peer high-altitude map;
step A4: point cloud data acquired by a laser radar;
step A5: downsampling the point cloud data by adopting a voxel filtering method;
step A6: performing ground segmentation and removal on the down-sampled point cloud data;
step A7: searching an initial pose value by adopting a branch-and-bound method based on the point cloud data and the occupied grid map;
step A8: and determining the maximum occupation probability by adopting a bilinear interpolation method, and optimizing the initial pose value by using a CERES optimizer.
Example two
Based on the same inventive concept as the above method, referring to fig. 3, this embodiment provides a global positioning device based on a contour map, including:
an acquisition module 31, configured to acquire first point cloud data and GNSS data acquired by a sensor;
a ground point removing module 32, configured to remove the point cloud ground points in the first point cloud data to obtain second point cloud data;
a map gridding module 33, configured to load a contour map of the region of interest using GNSS data, and gridding the contour map to obtain an occupied grid map;
An initial pose value obtaining module 34, configured to search a pose of the vehicle within a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
the initial pose value optimizing module 35 is configured to obtain, as a final pose, a pose occupying the grid map with the largest probability of being occupied by the second point cloud data, and optimize the initial pose value to obtain a final global positioning result.
Specifically, the ground point removal module 32 includes: the ordering unit is used for ordering the point cloud points of the first point cloud data according to the Z-axis value; the point cloud point removing unit is used for removing point cloud points with the Z-axis numerical value larger than the installation height of the sensor; the average value calculation unit is used for selecting a preset first number of point cloud points from small to large based on the Z-axis numerical value and calculating an average value; the initial point determining unit is used for traversing all the remaining point cloud points, calculating the difference value between the Z-axis value and the average value of the point cloud points, sorting the point cloud points with the difference value smaller than a preset threshold value from small to large, and selecting a preset second number of point cloud points as initial points of the point cloud ground; and the point cloud ground point removing unit is used for carrying out plane fitting on the initial points and removing the point cloud points on the plane through a fitted plane equation so as to finish the removal of the point cloud ground points.
The map gridding module 33 includes: the contour map acquisition unit is used for acquiring an initial value of the current positioning of the vehicle based on the GNSS data and determining a contour map according to the initial value; the contour map grid initializing unit is used for rasterizing the contour map and initializing the value of each grid unit in the grid map; a grid point determining unit for determining all grid points passed by the starting point position and the end point position of the contour line by using a light ray projection method; the grid unit value calculation unit is used for adding the values of the grid units according to the height of the current contour line for each passing grid point by using a preset value until all contour lines in the contour line map are traversed; the contour map normalization processing unit is used for normalizing the value of each grid unit in the grid map to obtain a normalized grid map serving as an occupied grid map.
The initial pose value acquisition module 34 includes: the recursion searching unit adopts a branch delimitation method to create a searching tree for calculating the pose of the vehicle, recursively searches a transformation matrix corresponding to each node on the searching tree, and multiplies the transformation matrix with the second point cloud data to obtain third point cloud data; the grid coordinate calculation unit is used for calculating grid coordinates corresponding to each point cloud point in the third point cloud data; the grid coordinate judging unit is used for judging whether the grid coordinates are in the range of occupying the grid map, if so, taking the value of the grid unit occupying the grid map as the score of the point cloud point, and if not, returning a negative value as the score of the point cloud point; the optimal transformation matrix determining unit is used for accumulating the scores of all the points and cloud points to obtain a score item, calculating whether the score item is an optimal score, if so, taking the transformation matrix obtained by current search as the optimal transformation matrix, and if not, pruning the nodes lower than the score item on the search tree; and the global positioning initial pose value determining unit is used for converting the second point cloud data into a world coordinate system according to the optimal transformation matrix so as to determine the initial pose value of the global positioning of the vehicle.
The initial pose value optimization module 35 includes: a CERES optimizer constructing unit for constructing a CERES optimizer and importing the second point cloud data and the occupation grid map into the CERES optimizer; an interpolation optimizing unit, configured to define an interpolator in the CERES optimizer, and interpolate each grid unit in the occupied grid map through the interpolator; the maximum occupation probability calculation unit is used for calculating the matching degree of the interpolated occupation grid map and the second point cloud data so as to determine the maximum occupation probability from the second point cloud data to the occupation grid map; and the maximum occupation probability optimizing unit is used for registering the second point cloud data and the contour map by a nonlinear optimizing method based on the maximum occupation probability so as to optimize the initial pose value.
The device also comprises: a downsampling processing module (not shown in the figure) for downsampling the first point cloud data, where the downsampling processing module specifically includes: the coordinate system conversion unit is used for multiplying the first point cloud data with a pre-calibrated sensor external parameter to obtain a laser point cloud under a vehicle body coordinate system; the voxel grid dividing unit is used for dividing a space area covered by the laser point cloud into a plurality of voxel grids and dividing points of the laser point cloud into corresponding voxel grids; and the sparse point cloud determining unit is used for representing all the point cloud points in the voxel grids through the mass centers of each voxel grid so as to construct sparse point cloud data.
The implementation process of the functions and actions of each module in the above device is specifically detailed in the implementation process of the corresponding steps in the above method, so relevant parts only need to be referred to in the description of the method embodiments, and are not repeated here.
The above-described embodiment of the apparatus is merely illustrative, for example, the division of the modules is merely a logic function division, and there may be another division manner in actual implementation, and each functional module in the embodiment may be all integrated in one processor, or each module may be separately used as one device, or two or more modules may be integrated in one device; the functional modules in the embodiments may be implemented in the form of hardware or in the form of hardware and software functional units.
Example III
Referring to fig. 4, the present embodiment provides an electronic apparatus including: processor 310 (processor), communication interface 320 (Communications Interface), memory 330 (memory) and communication bus 340, wherein processor 310, communication interface 320, memory 330 complete communication with each other through communication bus 340. The processor 310 may invoke logic instructions in the memory 330, where the processor 310 performs the global positioning method based on a contour map described in the above method embodiment, and the method includes:
Acquiring first point cloud data and GNSS data acquired by a sensor;
removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
loading a contour map of the region of interest by using GNSS data, and meshing the contour map to obtain an occupied grid map;
searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result.
Further, the logic instructions in the memory 330 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions to cause a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, where the computer program product includes a computer program, where the computer program can be stored on a non-transitory computer readable storage medium, where the computer program when executed by a processor can perform the global positioning method based on a contour map according to the above method embodiment, and the method includes:
acquiring first point cloud data and GNSS data acquired by a sensor;
removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
loading a contour map of the region of interest by using GNSS data, and meshing the contour map to obtain an occupied grid map;
searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result.
Example IV
The present embodiment provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the global positioning method based on a contour map described in the foregoing method embodiment, and the method includes:
Acquiring first point cloud data and GNSS data acquired by a sensor;
removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
loading a contour map of the region of interest by using GNSS data, and meshing the contour map to obtain an occupied grid map;
searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to serve as a final global positioning result.
The embodiments of the present invention are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the apparatus and medium embodiments, the description is relatively simple, as it is substantially similar to the method embodiments, with reference to the section of the method embodiments being relevant.
The devices and media provided in the embodiments of the present invention are in one-to-one correspondence with the methods, so that the devices and media also have similar beneficial technical effects as the corresponding methods, and since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media are not repeated here.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process article or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process article or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process method article or method comprising the element.
The foregoing is merely exemplary of the present invention and is not intended to limit the present invention. While the invention has been described in detail in the foregoing general description and with reference to specific embodiments thereof, it will be apparent to one skilled in the art that modifications and improvements can be made thereto. Accordingly, such modifications or improvements may be made without departing from the spirit of the invention and are intended to be within the scope of the invention as claimed.
Claims (10)
1. A global positioning method based on a contour map, comprising:
acquiring first point cloud data and GNSS data acquired by a sensor;
Removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
loading a contour map of the region of interest by using the GNSS data, and gridding the contour map to obtain an occupied grid map;
searching the vehicle pose in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
and acquiring the pose with the largest probability of occupying the grid map by the second point cloud data, and optimizing the initial pose value to obtain a final global positioning result.
2. The global positioning method based on a contour map as defined in claim 1, further comprising, prior to the step of removing the point cloud ground points in the first point cloud data: the downsampling process for the first point cloud data specifically includes:
multiplying the first point cloud data with a pre-calibrated sensor external parameter to obtain third point cloud data under a vehicle body coordinate system;
dividing the space region covered by the third point cloud data into a plurality of voxel grids, and dividing the points of the third point cloud data into corresponding voxel grids;
and representing all the point cloud points in the voxel grids through the centroid of each voxel grid so as to construct sparse point cloud data.
3. The global positioning method based on a contour map as defined in claim 1, wherein said step of removing point cloud ground points in said first point cloud data comprises:
ordering the point cloud points of the first point cloud data according to the Z-axis numerical value;
removing point cloud points with the Z-axis numerical value larger than the installation height of the sensor;
selecting a preset first number of point cloud points in sequence from small to large based on the Z-axis value and calculating a mean value;
traversing all the remaining point cloud points, calculating the difference value between the Z-axis value of the point cloud points and the mean value, sorting the point cloud points with the difference value smaller than a preset threshold value from small to large, and selecting a preset second number of point cloud points from the point cloud points as initial points of the point cloud ground;
and carrying out plane fitting on the initial points, and removing the point cloud points on the plane through a fitted plane equation to finish the removal of the point cloud ground points.
4. The global positioning method based on a contour map according to claim 1, characterized in that said loading a contour map of a region of interest using said GNSS data, gridding said contour map to obtain a step of occupying a grid map, comprises:
acquiring an initial value of the current positioning of the vehicle based on the GNSS data, and determining a contour map according to the initial value;
Rasterizing the contour map, and initializing the value of each grid unit in the grid map;
determining all grid points passed by the starting point position and the end point position of the contour line by using a light projection method, and adding the values of the grid units according to the current height of the contour line for each passed grid point by using a preset value until all contour lines in the contour line map are traversed;
and carrying out normalization processing on the value of each grid unit in the grid map to obtain a normalized grid map serving as the occupied grid map.
5. The global positioning method based on the contour map as set forth in claim 1, wherein,
the step of searching the vehicle pose in the preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning comprises the following steps:
creating a search tree for calculating the pose of the vehicle by adopting a branch-and-bound method, recursively searching a transformation matrix corresponding to each node on the search tree, and multiplying the transformation matrix by the second point cloud data to obtain third point cloud data;
calculating grid coordinates corresponding to the point cloud points for each point cloud point in the third point cloud data, judging whether the grid coordinates are in the range of the occupied grid map, if so, taking the value of the grid unit of the occupied grid map as the score of the point cloud point, and if not, returning a negative value as the score of the point cloud point;
Adding the scores of all the point cloud points to obtain a score term, calculating whether the score term is an optimal score, if so, taking the transformation matrix obtained by current search as an optimal transformation matrix, and if not, pruning nodes lower than the score term on the search tree;
and converting the second point cloud data into a world coordinate system according to the optimal transformation matrix so as to determine an initial pose value of global positioning of the vehicle.
6. The global positioning method based on a contour map according to claim 1, wherein said step of obtaining a pose with a maximum probability of occupation of said occupancy grid map by said second point cloud data optimizes said initial pose value comprises:
using a CERES optimizer, constructing in the CERES optimizer a maximum occupancy probability of the second point cloud data to the occupancy grid map;
and based on the maximum occupancy probability, registering the second point cloud data and the contour map by a nonlinear optimization method so as to optimize the initial pose value.
7. The global positioning method based on a contour map as defined in claim 6, wherein said step of constructing said second point cloud data to said maximum occupancy probability of said occupancy grid map in said CERES optimizer using a CERES optimizer comprises:
Constructing a CERES optimizer, and importing the second point cloud data and the occupied grid map into the CERES optimizer;
and defining an interpolator in the CERES optimizer, interpolating each grid unit in the occupied grid map through the interpolator, and calculating the matching degree of the interpolated occupied grid map and the second point cloud data so as to determine the maximum occupation probability from the second point cloud data to the occupied grid map.
8. A global positioning device based on a contour map, comprising:
the acquisition module is used for acquiring the first point cloud data and the GNSS data acquired by the sensor;
the ground point removing module is used for removing the point cloud ground points in the first point cloud data to obtain second point cloud data;
the map gridding module is used for loading a contour map of the region of interest by using the GNSS data, and gridding the contour map to obtain an occupied grid map;
the initial pose value acquisition module is used for searching the pose of the vehicle in a preset range based on the second point cloud data and the occupied grid map to obtain an initial pose value of global positioning;
the initial pose value optimizing module is used for acquiring the pose with the largest probability of occupying the grid map by the second point cloud data as the final pose, and optimizing the initial pose value to serve as the final global positioning result.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the global positioning method based on a contour map as claimed in any of claims 1-7 when executing the program.
10. A non-transitory computer readable storage medium, having stored thereon a computer program, which when executed by a processor implements the global positioning method based on a contour map as claimed in any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750255.9A CN116972859A (en) | 2023-06-21 | 2023-06-21 | Global positioning method, device, equipment and storage medium based on contour map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750255.9A CN116972859A (en) | 2023-06-21 | 2023-06-21 | Global positioning method, device, equipment and storage medium based on contour map |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116972859A true CN116972859A (en) | 2023-10-31 |
Family
ID=88473946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310750255.9A Pending CN116972859A (en) | 2023-06-21 | 2023-06-21 | Global positioning method, device, equipment and storage medium based on contour map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116972859A (en) |
-
2023
- 2023-06-21 CN CN202310750255.9A patent/CN116972859A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110988912B (en) | Road target and distance detection method, system and device for automatic driving vehicle | |
| CN110658531B (en) | Dynamic target tracking method for port automatic driving vehicle | |
| CN111860493B (en) | Target detection method and device based on point cloud data | |
| CN110276834B (en) | Construction method of laser point cloud map, terminal and readable storage medium | |
| CN114051628B (en) | Method and device for determining target object point cloud set | |
| CN113267761B (en) | Laser radar target detection and identification method, system and computer readable storage medium | |
| CN115451983A (en) | Dynamic environment mapping and path planning method and device in complex scene | |
| CN112166457A (en) | Point cloud segmentation method and system and movable platform | |
| CN114485698B (en) | Intersection guide line generation method and system | |
| CN111316328A (en) | Method for maintaining lane line map, electronic device and storage medium | |
| CN115205803A (en) | Automatic driving environment sensing method, medium and vehicle | |
| CN115265519A (en) | Online point cloud map construction method and device | |
| Lee et al. | An obstacle classification method using multi-feature comparison based on 2D LIDAR database | |
| CN114049362A (en) | Transform-based point cloud instance segmentation method | |
| CN117036447A (en) | Indoor scene dense three-dimensional reconstruction method and device based on multi-sensor fusion | |
| CN116972859A (en) | Global positioning method, device, equipment and storage medium based on contour map | |
| CN112906519B (en) | Vehicle type identification method and device | |
| CN116168384A (en) | Point cloud target detection method and device, electronic equipment and storage medium | |
| CN113808196A (en) | Plane fusion positioning method and device, electronic equipment and storage medium | |
| Vatavu et al. | Environment perception using dynamic polylines and particle based occupancy grids | |
| CN115082562A (en) | External parameter calibration method, device, equipment, server and vehicle-mounted computing equipment | |
| KR102624644B1 (en) | Method of estimating the location of a moving object using vector map | |
| CN117968682B (en) | Dynamic point cloud removing method based on multi-line laser radar and inertial measurement unit | |
| Drulea et al. | An omnidirectional stereo system for logistic plants. Part 2: stereo reconstruction and obstacle detection using digital elevation maps | |
| EP4141472A1 (en) | Computing architecture of an autonomous vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |