Method and System for Autonomous Driving Global Position Estimation Based on Association of Static Features in Standard Definition Electronic Map

Abstract

The present invention discloses a method and system for global position estimation of autonomous driving based on the association of static features in a standard-definition electronic map. The method includes obtaining a standard-definition electronic map of the autonomous driving urban operation scenario, performing parsing and preprocessing to extract a set of candidate vehicle position nodes during autonomous driving, judging and screening the static feature boundaries within a preset range for the set of candidate vehicle position nodes, and extracting the description information of urban structured static features on the standard-definition electronic map after screening to construct a prior static feature environment description database. By inputting real-time three-dimensional point cloud data through vehicle real-time perception, extracting the description of urban structured static features, and obtaining real-time static feature environment description data. Associating the static feature data in the standard-definition electronic map with the real-time static feature environment description data sensed by the vehicle end to obtain the result with the highest similarity to the vehicle end real-time perception, so as to realize the global position estimation of the autonomous driving vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to a method and system for global position estimation of autonomous driving based on the association of static objects in a standard definition electronic map. Background Art

[0002] The navigation, decision-making, and planning of autonomous driving rely on the accurate estimation of the vehicle's position on the map. Therefore, positioning is one of the key issues in the field of autonomous driving. In an urban scenario, when facing positioning inaccuracies caused by satellite signal occlusion, multipath effects, etc., the autonomous positioning method based on a pre-built high-precision map is the current mainstream solution. However, in scenarios where the coverage of the pre-built high-precision map is missing or poorly maintained, the classic "map building first - positioning later" technical route faces challenges of extremely degraded performance. Using a standard definition electronic map to achieve global position estimation of the vehicle will provide a new route for the existing "light map" autonomous driving technology, thereby improving the applicable range and functional integrity of autonomous driving vehicles.

[0003] Firstly, the standard definition electronic map contains road network information of urban structured scenarios and is applicable to the operating scenarios of most autonomous driving vehicles; secondly, the standard definition electronic map contains static object information of urban structured scenarios, which has commonality and similarity with vehicle-end perception; finally, the standard definition electronic map has a wide coverage range, is convenient to obtain, and has a high standardization maturity. Therefore, the standard definition electronic map can be used for the "map building - positioning" technology of autonomous driving vehicles in the case of missing pre-built high-precision maps. However, the information accuracy and richness in the standard definition electronic map are significantly different from the real-time perception modality at the vehicle end. How to solve the cross-modal information association is the main problem faced by this technology; in addition, there is a significant inconsistency between the information freshness in the standard definition electronic map and the vehicle-end real-time perception. How to avoid the influence of real-time dynamic information on the accuracy of the vehicle-end perception - standard definition electronic map information association is also the main problem faced by this technology. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for global position estimation of autonomous driving based on the association of static objects in a standard definition electronic map. By analyzing the static object environment information in the standard definition electronic map, virtual point clouds with the same mode as the vehicle-end perception are simulated and generated within the map position candidates. Furthermore, by using the direction change trend characteristics of static objects in the standard definition electronic map for environmental description and matching association basis, and extracting the direction trend change characteristics of the corresponding static objects in the vehicle's real-time perception, the problem of global position estimation of the vehicle under the conditions of significant differences in data accuracy, dimension, and mode faced by using the standard definition electronic map for autonomous driving vehicle positioning is solved.

[0005] The present invention provides a method for global position estimation of autonomous driving based on the association of static objects in a standard definition electronic map, including:

[0006] Obtain a standard definition electronic map under the automatic driving urban operation scenario;

[0007] Parse and preprocess the standard definition electronic map to extract a set of candidate nodes for vehicle positions during automatic driving;

[0008] Judge and screen the static object boundaries within a preset range for the set of candidate nodes for vehicle positions, and after screening, extract the description information of urban structured static objects on the standard definition electronic map to construct a prior static object environment description database;

[0009] Input three-dimensional point cloud data in real time through vehicle real-time perception, extract the description of urban structured static objects, and obtain real-time static object environment description data;

[0010] Associate the similarity between the static object data in the standard definition electronic map and the real-time static object environment description data sensed by the vehicle terminal in real time, and obtain the result with the highest similarity to the vehicle terminal real-time perception to achieve the global position estimation of the automatic driving vehicle.

[0011] Preferably, the parsing and preprocessing of the standard definition electronic map includes:

[0012] Obtain the road network data corresponding to the standard definition electronic map, and parse the connection relationship of the nodes in the standard definition electronic map;

[0013] Filter nodes by selecting road attributes, and screen out the roads with passability for automatic driving vehicles as the set P' of vehicle drivable roads, and the set P' of vehicle drivable roads includes several original road network node coordinates p' (x, y);

[0014] For two adjacent road nodes p1′(x1, y1) and p2′(x2, y2) belonging to the same road, calculate the plane geometric distance and the difference in horizontal and vertical coordinates between the two road nodes respectively, and determine the number of interpolation nodes to be inserted according to the preset node interval;

[0015] According to the difference X between the abscissas of two adjacent road nodes 差值 and the difference Y between the ordinates 差值 and the number of interpolation nodes to be inserted, calculate the coordinate increment p3′(X 增量， Y 增量 ) of each interpolation point to be inserted;

[0016] Distribute the interpolation points to be inserted equidistantly in the horizontal and vertical directions between the two road nodes according to the coordinate difference between the two road nodes, the number of interpolation nodes to be inserted, and the corresponding coordinate increment, and traverse and interpolate each original road network node in the set P' of drivable roads in the electronic map to form a set of candidate nodes for vehicle positions in the electronic map ;

[0017] Among them, X 差值 = x2 - x1, Y 差值 = y2 - y1, X 增量 = X 差值 / number of nodes to be interpolated, Y 增量 = Y 差值 / number of nodes to be interpolated, N represents the number of road network nodes after interpolation; p1, p2,... p N represent the planar node coordinates corresponding to the 1st, 2nd,... Nth road nodes.

[0018] Preferably, the parsing and preprocessing of the standard definition electronic map include:

[0019] Obtain the road network data corresponding to the standard definition electronic map, parse the connection relationship of the nodes in the standard definition electronic map, and extract all static object boundary information nodes in the parsed standard definition electronic map;

[0020] Filter the nodes by selecting boundary attributes, and screen out the boundaries with target static objects. The boundary attributes include buildings, fences, barriers, and road-related boundary objects;

[0021] Divide according to the same attributes of the boundaries, and construct a static object set composed of static object boundary points ;

[0022] Among them, K is the number of static object boundaries included in the standard definition electronic map. The element Bj in the set B represents the set of nodes of static objects corresponding to the same attribute, j = 1, 2,..., K; ; , ,... represent the planar two-dimensional coordinates of the 1st, 2nd,... nth nodes in the static object in the electronic map.

[0023] Preferably, the judgment and screening of the static object boundaries within a preset range for the vehicle position candidate node set include:

[0024] Traverse the vehicle position candidate node set P, and use the current vehicle position candidate node as the center on the geographic coordinate system in the standard definition electronic map. Through the preset prior visible range distance parameter R P , with R P as the radius, screen and truncate the static object boundaries within the range;

[0025] Calculate the distances between the candidate nodes of the current vehicle position and the corresponding nodes of all static objects to obtain a distance matrix. Each row of the distance matrix represents a candidate node of the vehicle position, each column represents a corresponding node of a static object, and the elements in the distance matrix represent the target distances from the candidate nodes of the vehicle position to the corresponding nodes of the static objects;

[0026] Determine whether the target distances in the distance matrix are greater than the static object nodes within the preset prior visible range. When the target distances are greater than the static object nodes within the preset prior visible range, filter the corresponding nodes of the static objects with the target distances and perform temporary filtering in the current state to form a local prior candidate set formed by a single global candidate node of the vehicle position and the boundaries within the prior visible range.

[0027] Preferably, the extraction of the description information of the urban structured static objects on the standard definition electronic map after screening is used to construct a prior static object environment description database, including:

[0028] Taking the current global candidate node of the vehicle position as the center on the geographic coordinate system in the standard definition electronic map, and using rays with the same angular interval to divide the plane where the current global candidate node of the vehicle position is located to form multiple divided regions arranged in counterclockwise order, and scan the regions near the boundary of the current global candidate node of the vehicle position;

[0029] Calculate the intersection points of each angular ray and the boundary line of the static object in turn, and take the plane coordinates corresponding to the points as the discrete point cloud coordinates in the divided region to generate a discrete static object boundary point cloud set , and have a unique corresponding relationship with the global candidate coordinates of the vehicle ;

[0030] Calculate the direction vectors between the discrete static object boundary point clouds in turn and the preset observation direction vector , and calculate the included angle between the current boundary direction vector and the preset observation direction vector, and form a one-dimensional vector as the prior static object boundary representation to achieve the extraction of the direction trend of the static object boundary in the standard definition electronic map scene;

[0031] Among them, the current direction vector is calculated in the following way:

[0032] ;

[0033] Among them, represents the number of columns of the vector, that is, the number of plane division rays, and respectively represent the position coordinates of the nth and (n - 1)th discrete static object boundary point clouds, represents the direction vector from the (n - 1)th discrete static object boundary point cloud to the nth discrete static object boundary point cloud, x i and y i are the starting coordinates of the vector x n and y n are the ending coordinates of the vector and are also the position coordinates of the nth discrete static object boundary point cloud. Preferably, the real-time input of three-dimensional point cloud data through vehicle real-time perception and the extraction of the description of urban structured static objects to obtain real-time static object environment description data include:

[0034] Taking the current vehicle observation point as the center, according to the three-dimensional point cloud data input by vehicle real-time perception, screening is performed according to the same attribute categories as the static objects in the prior standard-definition electronic map to obtain a three-dimensional point cloud set S of real-time static object boundaries with the same attributes. Through projection on the XOY plane in the vehicle body coordinate system, the three-dimensional point cloud set S is subjected to three-dimensional coordinate conversion to two-dimensional coordinates to obtain an unordered plane point cloud set S' containing only x coordinates and y coordinates;

[0035] Taking the current vehicle observation point as the center, rays with the same angular interval are used to divide the XOY plane of the current vehicle body coordinate system to form a plurality of divided regions arranged in counterclockwise order;

[0036] For the point clouds in each divided region, by solving the plane average center of the point cloud set in the divided region, the average center coordinates are used as the equivalent point cloud in the region, and finally a real-time static object boundary plane equivalent input point cloud set is obtained;

[0037] The direction vectors between the real-time static object boundary plane equivalent input point clouds and the corresponding observation direction vectors are calculated in sequence, and the included angle between the current boundary direction vector and the observation direction vector is calculated to form a one-dimensional vector as the real-time static object boundary environment representation to realize the extraction of the static object boundary direction trend in the vehicle real-time perception scenario;

[0038] Among them, the current direction vector is calculated in the following way:

[0039] ;

[0040] Among them, represents the number of columns of the vector, that is, the number of plane division rays, and respectively represent the position vector coordinates of the nth and (n - 1)th real-time point clouds, represents the direction vector from the (n - 1)th real-time static object boundary equivalent input point cloud to the nth real-time static object boundary equivalent input point cloud, x n and y n is the vector The vector coordinates in the vehicle body coordinate system XOY plane centered on the current vehicle observation point, and at the same time, it is the position coordinate of the nth static object boundary equivalent input point cloud.

[0041] Preferably, the method of associating the similarity between the static object data in the standard definition electronic map and the real-time static object environment description data sensed by the vehicle terminal to obtain the result with the highest similarity to the real-time perception of the vehicle terminal to achieve the global position estimation of the autonomous vehicle includes:

[0042] By traversing each element in the prior static object environment description database, querying the result most similar to the real-time perception in the prior static object environment description database, the similarity calculation is implemented by using the n-norm method as follows:

[0043] ,

[0044] where D(i) represents the similarity between the input static object environment description vector of the real-time perception and the ith candidate in the prior static object environment description database, represents the reference vector corresponding to the static object description of the real-time perception, represents the ith static object description vector corresponding to each vehicle global position candidate in the database, and N is the total number of static object environment description vectors in the prior static object environment description database;

[0045] Compare the calculated similarity results, screen out the candidate result with the highest similarity to the input perception vector, and obtain the global position estimation of the vehicle by accessing the vehicle global position candidate coordinates corresponding to the optimal candidate result.

[0046] The present invention also provides an autonomous driving global position estimation system based on the association of static objects in a standard definition electronic map, including:

[0047] A data acquisition module for obtaining a standard definition electronic map in the autonomous driving urban operation scenario;

[0048] An analysis and processing module for analyzing and preprocessing the standard definition electronic map to extract a set of vehicle position candidate nodes during autonomous driving;

[0049] An electronic map extraction module, which is used to judge and screen the static object boundaries within a preset range for the set of candidate vehicle position nodes, and extract the urban structured static object description information on the standard-definition electronic map after screening to construct a prior static object environment description database;

[0050] A real-time perception extraction module, which is used to input three-dimensional point cloud data in real time through vehicle real-time perception, and extract the urban structured static object description to obtain real-time static object environment description data;

[0051] A position estimation module, which is used to associate the similarity between the static object data in the standard-definition electronic map and the real-time static object environment description data sensed by the vehicle end, and obtain the position candidate result with the highest similarity to the vehicle end real-time perception to realize the global position estimation of the autonomous driving vehicle.

[0052] The present invention also provides an electronic device, including:

[0053] A memory, which is used to store a processing program;

[0054] A processor, when the processor executes the processing program, it realizes the autonomous driving global position estimation method based on the association of static objects in the standard-definition electronic map as described in the embodiments of the present invention.

[0055] The present invention also provides a readable storage medium, on which a processing program is stored, and when the processing program is executed by a processor, it realizes the autonomous driving global position estimation method based on the association of static objects in the standard-definition electronic map as described in the embodiments of the present invention.

[0056] Aiming at the prior art, the present invention has the following beneficial effects:

[0057] The present invention proposes an autonomous driving global position estimation method based on the association of static objects in the standard-definition electronic map. By analyzing the static object environment information in the standard-definition electronic map, virtual point clouds with the same mode as the vehicle end perception are simulated and generated within the map position candidates. Furthermore, it is proposed to use the direction change trend characteristics of static objects in the standard-definition electronic map as the basis for environmental description and matching association. By extracting the direction trend change characteristics of the corresponding static objects in the vehicle real-time perception, the problem of vehicle global position estimation under the conditions of significant differences in data accuracy, dimension, and mode faced by the standard-definition electronic map for autonomous driving vehicle positioning is solved. The invention has the following remarkable advantages and effects:

[0058] 1. It can get rid of the dependence on pre-built high-precision maps

[0059] The existing technical route of autonomous driving "mapping - positioning" relies on the accurate description of the static ground object environment provided in the pre - built high - precision map. The key to realizing positioning lies in the matching and registration of real - time perception and high - precision map elements. However, the method of the present invention directly obtains the static ground object environment information from a standard - definition electronic map and can be associated with the real - time perception at the vehicle end to achieve global position estimation, without the need to use a pre - built high - precision map.

[0060] 2. Wide range of applications

[0061] The existing technical route of autonomous driving "mapping - positioning" relies on a pre - built high - precision map, but the coverage of the pre - built high - precision map is limited. The standard - definition electronic map has a wide coverage and is easy to obtain. For areas where the pre - built map coverage is incomplete and the maintenance is lagging, the method of the present invention can effectively support the global positioning requirements of autonomous driving in areas where the pre - built map is missing.

[0062] 3. Cost reduction

[0063] The positioning route based on the pre - built high - precision map needs to maintain the freshness of map information through high - frequency maintenance to avoid the degradation of positioning performance. The existing map maintenance requires a high - precision map data collection vehicle and relies on manual annotation, with low automation. The positioning technical route of the present invention does not rely on a pre - built high - precision map, thus reducing the usage cost of the positioning solution.

[0064] In summary, through theoretical analysis and experimental data, the present invention proves that the standard - definition electronic map has the potential and technical feasibility to replace the pre - built high - precision map in the positioning of urban structured scenarios. At the same time, it significantly improves the accuracy of using the standard - definition electronic map to provide vehicle global position estimation, reduces the promotion cost of the autonomous driving "mapping - positioning" technical route, and has important practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic diagram of the steps of the autonomous driving global position estimation method based on the association of static ground objects in the standard - definition electronic map according to an embodiment of the present invention;

[0066] Figure 2 It is a principle block diagram of the autonomous driving global position estimation system based on the association of static ground objects in the standard - definition electronic map according to an embodiment of the present invention;

[0067] Figure 3 It is a schematic diagram of the prior static ground object environment description according to an embodiment of the present invention;

[0068] Figure 4 It is a schematic diagram of the real - time static ground object environment description according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0070] As used herein, the term "including" and its variants are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "an embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0071] It should be noted that the concepts such as "first" and "second" mentioned in the disclosure of this application are only used to distinguish different devices, modules, or units, and are not used to limit the order or mutual dependence relationship of the functions performed by these devices, modules, or units.

[0072] It should be noted that the modifications of "one" and "multiple" mentioned in the disclosure of this application are illustrative rather than restrictive. Those skilled in the art should understand that unless clearly stated otherwise in the context, it should be understood as "one or more".

[0073] Embodiment 1

[0074] As Figures 1-4 shown, the present invention also provides an autonomous driving global position estimation method based on the association of static objects in a standard definition electronic map, which is applicable to an autonomous driving global position estimation system based on the association of static objects in a standard definition electronic map. The system consists of the following main modules: a standard definition electronic map parsing and preprocessing module, a city structured static object description extraction module for the standard definition electronic map, and a city structured static object description extraction module for real-time perception. The method includes the following steps:

[0075] S1: Obtain a standard definition electronic map in the autonomous driving urban operation scenario;

[0076] S2: Parse and preprocess the standard definition electronic map to extract a set of vehicle position candidate nodes during autonomous driving, including three steps: electronic map road network parsing, electronic map road network interpolation, and static object boundary parsing, which are described in detail as follows:

[0077] 1. Electronic map road network parsing: Obtain the road network data corresponding to the standard definition electronic map, and parse the connection relationship of the nodes in the standard definition electronic map. The information in the standard definition electronic map consists of points, lines, and polygons. Taking the eastward direction in the electronic map as the positive x-axis and the northward direction as the positive y-axis, the set P' of drivable roads consists of several original road network node coordinates p'(x, y).

[0078] By selecting road attributes for node filtering, that is, using attributes such as road grades and types in the map, filter out the roads with passability for autonomous vehicles as the set P' of vehicle drivable area roads. The set P' of vehicle drivable area roads includes several original road network node coordinates p'(x, y).

[0079] 2. Electronic map road network interpolation: Since the original road network nodes are relatively sparse. For example, the nodes in the electronic map are very unevenly distributed. If a 100-meter road is defined with only 3 nodes and a 10-meter road is defined with 5 nodes, interpolation operations are performed to ensure the uniformity of node distribution on each road. Specifically, by parsing the set of vehicle drivable area roads with vehicle passability, for two adjacent road nodes p1'(x1, y1) and p2'(x2, y2) belonging to the same road, calculate the planar geometric distance and the difference in horizontal and vertical coordinates between the two road nodes respectively, and determine the number of interpolation nodes to be taken as an integer according to the preset node interval;

[0080] According to the difference X between the abscissas of two adjacent road nodes 差值 and the difference Y between the ordinates 差值 and the number of interpolation nodes to be taken, calculate the coordinate increment p3'(X 增量， Y 增量 ) of each interpolation point to be taken;

[0081] Distribute the interpolation points to be taken equidistantly in the horizontal and vertical directions between the two road nodes according to the coordinate difference between the two road nodes, the number of interpolation nodes to be taken, and the corresponding coordinate increment. Traverse and interpolate each original road network node in the set P' of all drivable roads in the standard definition electronic map using the above method to form a set of candidate nodes for the global vehicle position in the electronic map ;

[0082] Among them, X 差值 = x2 - x1, Y 差值 = y2 - y1, X 增量 = X 差值 / the number of interpolation nodes to be taken, Y 增量 = Y 差值 / the number of interpolation nodes to be taken, N represents the number of interpolated road nodes in the electronic map; p1, p2,...p Nrepresent the planar node coordinates corresponding to the 1st, 2nd, ..., Nth road nodes. Through the above method, road network data can be effectively extracted and processed from a standard-definition electronic map, providing accurate road information for autonomous driving vehicles. Facing different position estimation requirements, this interval can be determined as needed. Additionally, the parameters of this interval can be determined according to the scene size. The smaller the interval, the larger the number of databases to be matched. By performing interpolation processing on the original road network nodes based on road curvature, the density of the road network can be effectively increased, thereby providing more accurate and rich navigation information for autonomous driving vehicles.

[0083] In this embodiment, the static object boundary parsing also includes parsing the static object boundaries in the map, which is described in detail as follows:

[0084] Obtain the road network data corresponding to the standard-definition electronic map, parse the connection relationships of the nodes in the standard-definition electronic map, and extract all the static object boundary information nodes in the parsed standard-definition electronic map;

[0085] By selecting boundary attributes for node filtering, filter out the nodes with target static object boundaries, and the boundary attributes include buildings, fences, barriers, and road-related boundary objects;

[0086] Divide according to the same attributes of the boundaries to construct a static object set composed of static object boundary points ;

[0087] where K is the number of static object boundaries included in the standard-definition electronic map, and the element Bj in the set B represents the set of nodes belonging to the static objects corresponding to the same attribute, j = 1, 2, ..., K; ; , ,... represent the planar two-dimensional coordinates of the 1st, 2nd, ..., nth nodes in the static object in the electronic map. By accurately parsing the connection relationships of the nodes in the standard-definition electronic map, it can be ensured that the extracted static object boundary information is accurate and error-free, providing a reliable basis for subsequent processing; using boundary attributes for node filtering and division can effectively distinguish different types of static object boundaries, improving the efficiency and accuracy of data processing; when constructing the static object set, dividing according to the same attributes of the boundaries helps to simplify the data structure and facilitate subsequent analysis and processing; by clearly representing the planar two-dimensional coordinates of each node in the static object in the electronic map, accurate environmental perception information can be provided for the autonomous driving system, improving driving safety and stability.

[0088] S3: Judge and screen the static object boundaries within a preset range for the vehicle position candidate node set, and extract the urban structured static object description information on the standard definition electronic map after screening to construct a prior static object environment description database; the following is a detailed description of the visible static object boundary screening step:

[0089] Traverse the vehicle global position candidate node set P. In the geographic coordinate system within the standard definition electronic map, such as the UTM / WGS84 coordinate system, with the current vehicle global position candidate node as the center, through the preset prior visible range distance parameter R P , with R P as the radius, screen and truncate the static object boundaries within the range; it can be understood that this visible static object boundary screening step uses the vehicle global position candidate node set P, by introducing the prior visible range distance parameter R P , with R P as the radius, screen and truncate the static object boundaries within the range, forming a local prior candidate set composed of a single candidate position node and the boundaries within the visible range; alternatively, preferably, screen out the nearest boundary points in each direction by simulating the light occlusion characteristics;

[0090] To remove the static object boundary nodes outside the visible range, calculate the distances between the current vehicle global position candidate node and all static object corresponding nodes to obtain a distance matrix. Each row of the distance matrix represents a vehicle global position candidate node, each column represents a static object corresponding node, and the elements in the distance matrix represent the target distances from the vehicle global position candidate node to the static object corresponding node;

[0091] Judge whether the target distances in the distance matrix are greater than the static object nodes within the preset prior visible range. When the target distances are greater than the static object nodes within the preset prior visible range, screen the static object corresponding nodes corresponding to the target distances and perform temporary filtering in the current state, forming a local prior candidate set composed of a single vehicle global position candidate node and the boundaries within the prior visible range. The temporary filtering is to generate the static object environment description corresponding to this position candidate node.

[0092] In this embodiment, the step of extracting the urban structured static object description information on the standard definition electronic map after screening in step S3 to construct a prior static object environment description database further includes: a planar static object boundary discrete point cloud generation step and a static object boundary direction trend extraction, which are described in detail as follows:

[0093] With the current vehicle global position candidate node as the center in the geographic coordinate system, take the same angular interval The rays divide the plane where the current vehicle's global position candidate nodes are located into multiple divided regions arranged in counterclockwise order, and scan the regions near the boundary of the current vehicle's global position candidate nodes; for example, after visual boundary screening, with the candidate position node as the center, take the same angular interval The rays divide the plane and perform a 360° counterclockwise circumferential scan on the boundary near the candidate point. Radiating outward with the candidate position as the center, the intersection point where the ray first intersects the static object boundary is the visual observation point. In the above and below embodiments, the division operation is to simulate the scanning mode of lidar. The roads, buildings and other static objects in the high-precision electronic map are described by polygons composed of discrete points, which is very different from the point cloud mode. In order to unify the expression method with the vehicle lidar point cloud, it is necessary to divide the plane and perform corresponding calculations.

[0094] Generation of discrete point cloud of plane static object boundary: Calculate the intersection points of each angular ray and the static object boundary line in turn, and take the plane coordinates corresponding to the points as the discrete point cloud coordinates in the divided region to generate a discrete static object boundary point cloud set and has a unique corresponding relationship with the vehicle's global position candidate coordinates That is, each global position candidate coordinate has a unique discrete static object boundary point cloud set corresponding to it;

[0095] Extraction of static object boundary direction trend: For the given plane virtual point cloud set s′ and its corresponding vehicle global position candidate at equal angular intervals starting from the positive x-axis in the counterclockwise direction, calculate the direction vectors between the discrete static object boundary point clouds in turn, as well as the preset observation direction vector , and calculate the included angle between the boundary direction vector and the preset observation direction vector. The preset observation direction here is the ray direction, that is, the observation direction of the laser during the simulation of lidar point cloud scanning. The formed one-dimensional vector is used as the prior static object boundary representation ;

[0096] Among them, the direction vector is calculated in the following way:

[0097] ;

[0098] Among them, represents the number of columns of the vector, that is, the number of plane division rays, that is, the number of sectors divided by the plane = 360 degrees of the circumference divided by the ray angle interval and round up, and respectively represent the position coordinates of the nth and (n - 1)th discrete static object boundary point clouds, represents the direction vector from the (n - 1)th discrete static object boundary point cloud to the nth discrete static object boundary point cloud, that is, the boundary direction vector corresponding to the nth ray. The calculation of this direction vector is from the point cloud coordinates corresponding to the (n - 1)th ray to the point cloud coordinates corresponding to the nth ray, x i and y i are the starting coordinates of the vector, x n and y n are the ending coordinates of the vector, and at the same time are the position coordinates of the nth discrete static object boundary point cloud. Among them, the number of rays = the number of planar point clouds = the number of direction vectors.

[0099] The principle of this embodiment is to use the polygons and lines in the standard definition electronic map to generate virtual point clouds in the same mode as the vehicle-end perception, and use the boundary directions extracted from the virtual point clouds to solve the problem of cross-modal information association. The vehicle-end perception itself is point cloud, and the boundary direction can be directly extracted. The specific implementation steps are as follows:

[0100] S4: Real-time input three-dimensional point cloud data through vehicle real-time perception, extract the description of urban structured static objects, and obtain real-time static object environment description data, including:

[0101] 1. Plane projection of real-time static object boundary point cloud: With the current vehicle observation point as the center, define the forward direction of the vehicle as the positive y-axis, the right side of the vehicle as the positive x-axis, and the top of the vehicle as the positive z-axis to construct a vehicle body coordinate system. For the dense three-dimensional point cloud input by vehicle-end real-time perception, filter according to the same attribute categories as the static objects in the prior standard definition electronic map, and obtain the real-time static object boundary three-dimensional point cloud set . Through the XOY plane projection, filter the elevation information of the point cloud to obtain an unordered plane point cloud set without elevation ;

[0102] 2. Generation of real-time static object plane equivalent point cloud:

[0103] With the current vehicle observation point as the center, in the counterclockwise direction as positive, take rays with the same angular interval to divide the plane where the current vehicle global position candidate node is located, forming multiple divided regions arranged in counterclockwise order;

[0104] For the point cloud in each divided region, by solving the plane average center of the point cloud set in this divided region, use the average center coordinates as the equivalent point cloud in this region, and finally obtain the real-time static object boundary plane equivalent input point cloud set ;

[0105] 3. Real-time static object boundary direction trend extraction: At the same angular interval , with counterclockwise being positive, sequentially calculate the direction vectors between the equivalent input point clouds of the real-time static object boundary plane and the corresponding observation direction vectors , and calculate the angle between the current boundary direction vector and the observation direction vector . Here, the observation direction is the direction formed by the measurement coordinates returned by the laser sensor, which is a one-dimensional vector as the real-time static object boundary environment description;

[0106] Among them, the current direction vector is calculated as follows:

[0107] ;

[0108] Among them, represents the number of columns of the vector, that is, the number of plane segmentation rays, and respectively represent the position vector coordinates of the nth and (n - 1)th real-time point clouds. The vector represents the direction vector from the (n - 1)th real-time static object boundary equivalent input point cloud to the nth real-time static object boundary equivalent input point cloud. x n and y n are the vector coordinates of the vector in the vehicle body coordinate system XOY plane centered on the current vehicle observation point, and at the same time are the position coordinates of the nth real-time static object boundary equivalent input point cloud.

[0109] S5: Associate the static object data in the standard definition electronic map with the real-time static object environment description data sensed by the vehicle end in real time, and obtain the result with the highest similarity to the real-time perception at the vehicle end to achieve the global position estimation of the autonomous vehicle. This step is implemented by the vehicle global position estimation system. This system includes a prior static object environment description database generated from the standard definition electronic map. Each element in this database is composed of a row vector of 1x dimensions, corresponding to each vehicle candidate position in the standard definition electronic map; the real-time static object environment description is composed of a row vector of 1x dimensions generated by inputting each frame of point cloud;

[0110] By traversing each element in the prior static object environment description database, query the result most similar to the real-time perception in the prior static object environment description database. The similarity calculation is implemented using the n-norm method as follows: ,

[0111] Among them, D(i) represents the similarity between the input vector and the ith candidate in the prior static object environment description database. It represents the reference vector corresponding to the static object description for real-time perception. It represents the i-th static object description vector corresponding to each vehicle global position candidate in the database. N is the total number of vectors in the database. Generally, n = 1 or n = 2 is taken, that is, the 1-norm or 2-norm is used for vector similarity calculation. The method of this embodiment is effective under different values, only having a slight impact on the calculation efficiency.

[0112] Compare the calculated similarity results, screen out the candidate result with the highest similarity to the input perception vector, and obtain the global position estimate of the vehicle by accessing the vehicle global position candidate coordinates corresponding to the optimal candidate result. Through the above specific implementation manners, the present invention can use the static object information in the standard definition electronic map to be associated with the vehicle real-time perception to realize the global position estimate of the autonomous driving vehicle, getting rid of the dependence on the pre-built high-precision map.

[0113] In the embodiment of the present invention, by parsing the static object environment information in the standard definition electronic map, virtual point clouds in the same mode as the vehicle-end perception are simulated and generated within the map position candidates, and further, a method is proposed to use the direction change trend characteristics of the static objects in the standard definition electronic map as the basis for environment description and matching association. By extracting the direction trend change characteristics of the corresponding static objects in the vehicle real-time perception, the problem of global position estimation of the vehicle under the conditions of significant differences in data accuracy, dimension, and mode faced by the standard definition electronic map for autonomous driving vehicle positioning is solved. In the experiment, the technical solution proposed in the embodiment of the present invention was compared with the existing methods in the autonomous driving data set of the urban structured scene. The results show that in the urban scene sequence of KITTI, the present invention has better performance in terms of the positioning success rate within 2 meters and the positioning success rate within 5 meters. And under the conditions of different confidence levels for obtaining real-time boundary point clouds, the embodiment of the present invention shows higher robustness and accuracy.

[0114] The specific experimental data is as follows: In the typical sequence of the KITTI urban structured scene, when using a large-scale standard definition electronic map (including 11,612 position candidates), the positioning success rate within 2 meters is 43.27%, and the positioning success rate within 5 meters is 53.20%; when using a small-scale standard definition electronic map (including 4,541 position candidates), the positioning success rate within 5 meters is 57.8%. It is significantly better than other existing positioning methods.

[0115] In summary, the embodiment of the present invention proves through theoretical analysis and experimental data that the standard definition electronic map has the potential and technical feasibility to replace the pre-built high-precision map in the urban structured scene in terms of positioning. At the same time, it significantly improves the accuracy of using the standard definition electronic map to provide the global position estimate of the vehicle, reduces the promotion cost of the "map building - positioning" technical route for autonomous driving, and has important practical application value.

[0116] Example Two

[0117] Based on the same concept, the present invention provides an autonomous driving global position estimation system based on the association of static objects in a standard-definition electronic map, including:

[0118] A data acquisition module, configured to obtain a standard-definition electronic map of the autonomous driving urban operation scenario;

[0119] An analysis and processing module, i.e., a standard-definition electronic map analysis and preprocessing module, configured to analyze and preprocess the standard-definition electronic map to extract a set of candidate vehicle position nodes during autonomous driving;

[0120] An electronic map extraction module, i.e., an urban structured static object description extraction module for the standard-definition electronic map, configured to judge and screen the static object boundaries within a preset range for the set of candidate vehicle position nodes, and extract the urban structured static object description information on the standard-definition electronic map after screening to construct a prior static object environment description database;

[0121] A real-time perception extraction module, i.e., an urban structured static object description extraction module for real-time perception, configured to input three-dimensional point cloud data in real time through vehicle real-time perception, extract the urban structured static object description, and obtain real-time static object environment description data;

[0122] A position estimation module, configured to perform similarity association between the static object data in the standard-definition electronic map and the real-time static object environment description data sensed by the vehicle end in real time, and obtain a position candidate result with the highest similarity to the real-time perception of the vehicle end to achieve the global position estimation of the autonomous driving vehicle. The system is specifically described as follows:

[0123] 1. Standard-definition electronic map analysis and preprocessing module

[0124] 1.1 Electronic map road network analysis:

[0125] The information in the standard-definition electronic map is composed of points, lines, and polygons. By analyzing the connection relationships of the nodes in the standard-definition electronic map and filtering using attributes such as road grades and types in the map, the roads with theoretical passability for autonomous driving vehicles are retained as the set of vehicle drivable roads.

[0126] 1.2 Electronic map road network interpolation:

[0127] By parsing the set of vehicle - drivable road areas where vehicles can pass, for two adjacent road nodes belonging to the same road, calculate the planar geometric distances and horizontal and vertical coordinate interpolations of the two nodes in the horizontal and vertical directions respectively. Calculate and round up the number of interpolation nodes to be calculated according to the preset road node interval, and distribute the interpolation points equidistantly in the horizontal and vertical directions between the two road nodes to form a set of candidate nodes for the global vehicle position in the electronic map.

[0128] 1.3 Static feature boundary parsing:

[0129] By parsing all static feature boundary information nodes in the standard - definition electronic map, filter through boundary attributes (such as buildings, fences, barriers, etc.) to retain static feature boundaries with long - term validity; secondly, further divide according to the co - ownership nature of the boundaries to construct a set of static objects composed of static feature boundary points.

[0130] 2. Urban structured static feature description extraction module for standard - definition electronic maps

[0131] 2.1 Visible boundary screening:

[0132] Using the set of candidate nodes for the global vehicle position generated by module 1.2, filter the static feature boundaries within the visible range of the candidate nodes through the preset prior visible range distance parameter to form a local prior candidate set composed of a single candidate position node and the boundaries within the visible range.

[0133] 2.2 Generation of planar boundary discrete point cloud:

[0134] After visible boundary screening, with the candidate position node as the center, the east direction of the map as the x - axis, the north direction of the map as the y - axis, and counter - clockwise as positive, use rays at preset equal - angle intervals to divide the plane. Scan the boundary near the candidate point counter - clockwise for 360°. Calculate the planar coordinates of the intersection points of the scanning rays and the static feature boundary lines in turn as the point cloud coordinates in the divided area, and then generate a set of prior local boundary discrete point clouds, which together with the vehicle candidate position coordinates form a corresponding local planar point cloud virtual observation data set.

[0135] 2.3 Extraction of static feature boundary direction trend:

[0136] Based on the planar virtual point cloud observations generated by module 2.2, at the same angle interval as in 2.2, in counter - clockwise order, starting from the positive x - axis, calculate the direction vectors between discrete static feature boundary point clouds and the preset observation direction vectors in turn, and further calculate the included angle between the boundary direction vector and the preset observation direction vector. The formed one - dimensional vector is used as the prior static feature environment description.

[0137] 3. Urban structured static feature description extraction module for real - time perception

[0138] 3.1 Real-time static object boundary point cloud planar projection:

[0139] For the dense three-dimensional point cloud input by real-time perception at the vehicle end, filter it according to categories with the same attributes as the static object boundaries in the prior map, such as buildings, fences, retaining walls, etc., to obtain a real-time boundary three-dimensional point cloud set. Taking the vehicle observation point as the center, defining the vehicle forward direction as the positive y-axis direction, the vehicle right side as the positive x-axis direction, and the vehicle top as the positive z-axis direction, through the XOY plane projection, filter the Z-axis information of the point cloud to obtain an unordered planar point cloud set without elevation;

[0140] 3.2 Real-time static object planar equivalent point cloud generation:

[0141] Taking the vehicle observation point as the center, defining the vehicle forward direction as the positive y-axis direction, the vehicle right side as the positive x-axis direction, with counterclockwise being positive, divide the planar area at the same angular interval. For the planar point cloud within each divided area, determine the center point by calculating the average value of the coordinates of the point cloud set within the area in the x-axis and y-axis directions, and use the center point coordinates as the equivalent point cloud within the area. The real-time planar segmentation quantity and angular interval are the same as the configuration in Module 2.2;

[0142] 3.3 Real-time static object boundary direction trend extraction:

[0143] At the same angular interval, with counterclockwise being positive, sequentially calculate the direction vectors between discrete boundary point clouds and the corresponding observation direction vectors, and further calculate the included angle between the boundary direction vector and the observation direction vector. The formed one-dimensional vector is used as the real-time static object environment description.

[0144] 4. Implementation of vehicle global position estimation system

[0145] This system includes a prior static object description database generated from a standard definition electronic map. Each element in this database consists of several row vectors, corresponding to each vehicle candidate position in the standard definition electronic map. The number of database elements depends on the range size of the used standard definition electronic map, with no specific quantity limit; the real-time static object environment description consists of static object description vectors generated corresponding to each frame of point cloud input; by traversing each element in the prior static object description database, a result most similar to the real-time perception can be found in the prior static object description database, and by accessing the position coordinates corresponding to this result, the global position estimation of the vehicle can be obtained.

[0146] It should be noted that the division of each module in the embodiments of the present device / system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or physically separated. And these modules can all be implemented in the form of software called by a processing element; they can also all be implemented in the form of hardware; or some units can be implemented in the form of software called by a processing element, and some units can be implemented in the form of hardware.

[0147] The implementation principles of the above modules have been described in the foregoing embodiments, so they will not be repeated here.

[0148] Embodiment III

[0149] Based on the same concept, in some embodiments of the present application, an electronic device is further provided. This electronic device includes a memory and a processor, where the memory is used to store a processing program, and the processor executes the processing program according to instructions. When the processor executes the processing program, the method for estimating the global position of an autonomous vehicle based on the association of static objects in a standard-definition electronic map in the foregoing embodiments is realized.

[0150] In some embodiments of the present application, a readable storage medium is further provided. This readable storage medium can be a non-volatile readable storage medium or a volatile readable storage medium. Instructions are stored in this readable storage medium. When these instructions run on a computer, an electronic device including this readable storage medium is caused to execute the foregoing method for estimating the global position of an autonomous vehicle based on the association of static objects in a standard-definition electronic map.

[0151] It can be understood that for the foregoing method for estimating the global position of an autonomous vehicle based on the association of static objects in a standard-definition electronic map, if it is all implemented in the form of software functional modules and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present invention. And the foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0152] A computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable storage medium may also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0153] The program code for implementing the technical solutions disclosed in this application can be written in any combination of one or more programming languages. The programming languages include object-oriented programming languages such as Python, C++, etc., and also include conventional procedural programming languages such as C language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for global position estimation of autonomous driving based on static object association of standard definition electronic map, characterized in that: include: Obtain standard-definition electronic maps for autonomous driving city operation scenarios; The standard definition electronic map is parsed and preprocessed to extract a set of candidate nodes for the vehicle position during autonomous driving, including: obtaining road network data corresponding to the standard definition electronic map, parsing the connection relationship of the nodes in the standard definition electronic map; filtering the nodes by selecting road attributes, and screening out roads that are passable by autonomous driving vehicles as a set of roads in the vehicle's drivable domain, wherein the set of roads in the vehicle's drivable domain includes a number of original road network node coordinates p' (x, y); for two adjacent road nodes p1' (x1, y1) and p2' (x2, y2) belonging to the same road, respectively calculating the plane between the two road nodes The geometric distance and the difference of the horizontal and vertical coordinates are calculated, and the number of nodes to be interpolated is determined according to the preset node interval; the coordinate increment p3′ (X increment, Y increment) of each point to be interpolated is calculated according to the difference X between the horizontal coordinates of two adjacent road nodes, the difference Y between the vertical coordinates and the number of nodes to be interpolated; the points to be interpolated are distributed at equal distances in the horizontal and vertical directions between the two road nodes according to the coordinate difference of the two road nodes, the number of nodes to be interpolated and the corresponding coordinate increments; each original road network node in the set P' of all drivable domain roads in the standard definition electronic map is traversed and interpolated according to the above method to form a set of candidate vehicle position nodes in the electronic map. ; Wherein, X difference = x2-x1, Y difference = y2-y1, X increment = X difference / number of nodes to be interpolated, Y increment = Y difference / number of nodes to be interpolated, N represents the number of road nodes after interpolation in the electronic map; p1, p2, ... pN represent the plane node coordinates corresponding to the 1st, 2nd, ... Nth road nodes; The static object boundaries within a preset range are judged and screened for the candidate node set of the vehicle position, and after the screening, the description information of the structured static objects in the city on the standard definition electronic map is extracted to construct a priori static object environment description database; Through real-time vehicle perception and real-time input of 3D point cloud data, the description of urban structured static objects is extracted to obtain real-time static object environment description data; The static object data in the standard definition electronic map is similarly associated with the real-time static object environment description data perceived by the vehicle side, and the position candidate result with the highest similarity to the real-time perception of the vehicle side is obtained to realize the global position estimation of the autonomous driving vehicle, including: by traversing each element in the prior static object environment description database, searching the prior static object environment description database for the result most similar to the real-time perception, and the similarity calculation is realized by using the n-norm method as follows: , where D(i) represents the similarity between the input static object environment description vector perceived in real time and the i-th candidate in the prior static object environment description database, Represents the reference vector corresponding to the real-time perceived static feature description, represents the i-th static object description vector corresponding to each vehicle global position candidate in the database, and N is the total number of static object environment description vectors in the prior static object environment description database; the calculated similarity results are compared to screen out the candidate result with the highest similarity to the input perception vector, and the vehicle global position estimation is obtained by accessing the vehicle global position candidate coordinates corresponding to the optimal candidate result; The parsing and preprocessing of the standard definition electronic map includes: Obtaining the road network data corresponding to the standard definition electronic map, parsing the connection relationship of the nodes in the standard definition electronic map, and extracting all the static feature boundary information nodes in the parsed standard definition electronic map; Filter nodes by selecting boundary attributes to select target static objects with boundary, wherein the boundary attributes include buildings, fences, barriers and road-related boundary objects; Divide according to the same attributes of the boundary and construct a static object set composed of static ground feature boundary points ; Where K is the number of static feature boundaries contained in the standard definition electronic map, and the element B in the set B is j Represents the set of nodes belonging to static objects corresponding to the same attribute, j=1, 2, ..., K; ; , ,... Represents the two-dimensional coordinates of the 1st, 2nd, ...nth nodes in the static feature in the electronic map.

2. The method for global position estimation of autonomous driving based on static object association of standard definition electronic map according to claim 1, characterized in that: The determining and screening of the static object boundaries within a preset range of the vehicle position candidate node set includes: Traverse the set of candidate vehicle position nodes P, take the current vehicle global position candidate node as the center in the geographic coordinate system of the standard definition electronic map, and use the preset prior visual range distance parameter R P , with R P The static features within the radius are screened and truncated; Calculate the distance between the current vehicle position candidate node and all the static object corresponding nodes to obtain a distance matrix, wherein each row of the distance matrix represents a vehicle position candidate node, each column represents a static object corresponding node, and the elements in the distance matrix represent the target distance from the vehicle position candidate node to the static object corresponding node; Determine whether the target distance in the distance matrix is ​​greater than the static object node within the preset prior visible range. When the target distance is greater than the static object node within the preset prior visible range, screen the static object corresponding nodes corresponding to the target distance and perform temporary filtering in the current state to form a local prior candidate set formed by a single vehicle global position candidate node and a boundary within the prior visible range.

3. The method for global position estimation of autonomous driving based on static object association of standard definition electronic map according to claim 1, characterized in that: The method of extracting the city structured static land object description information on the standard definition electronic map after screening to construct a priori static land object environment description database includes: In the geographic coordinate system of the standard definition electronic map, the candidate node of the current vehicle global position is taken as the center of the circle, and the same angle interval is taken. The ray of the current vehicle global position candidate node is segmented to form a plurality of segmented areas arranged in a counter-clockwise order, and the area close to the boundary of the current vehicle global position candidate node is scanned; Calculate the points where each angle ray intersects with the boundary line of the static object in turn, and take the plane coordinates corresponding to the point as the discrete point cloud coordinates in the segmented area to generate a discrete static object boundary point cloud set , and the candidate coordinates of the vehicle global position Have a unique corresponding relationship; Calculate the direction vectors between discrete static object boundary point clouds in sequence And the preset observation direction vector , and calculate the angle between the current boundary direction vector and the preset observation direction vector The one-dimensional vector formed is used as a priori static feature boundary representation , in order to realize the extraction of the boundary direction trend of static objects in the standard definition electronic map scene; Among them, the current direction vector is calculated by the following method: ； in, Indicates the number of columns of the vector, i.e. the number of plane splitting rays, and Respectively represent the position coordinates of the nth and n-1th discrete static object boundary point clouds, Indicates the direction vector from the n-1th discrete static object boundary point cloud to the nth discrete static object boundary point cloud, x i and i For vector The starting point coordinates, x n and n For vector The end point coordinates are also the position coordinates of the nth discrete static feature boundary point cloud.

4. The method for global position estimation of autonomous driving based on static object association of standard definition electronic map according to claim 1, characterized in that: The real-time input of three-dimensional point cloud data by vehicle real-time perception to extract the description of urban structured static objects and obtain real-time static object environment description data includes: Taking the current vehicle observation point as the center, based on the 3D point cloud data input by the vehicle-side real-time perception, the data is screened according to the same attribute categories as the static objects in the prior standard definition electronic map, and a real-time static object boundary 3D point cloud set S with the same attributes is obtained. The 3D point cloud set S is converted into 2D coordinates by XOY plane projection in the vehicle body coordinate system, and an unordered plane point cloud set S′ containing only x-coordinates and y-coordinates is obtained. Centered on the current vehicle observation point, take the same angle interval The ray divides the XOY plane of the current vehicle coordinate system into multiple divided areas arranged in a counter-clockwise order; For each point cloud in the segmented area, the plane average center of the point cloud set in the segmented area is solved, and the coordinates of the average center are used as the equivalent point cloud in the area, and finally the real-time static feature boundary plane equivalent input point cloud set is obtained. ; Calculate the direction vectors between the equivalent input point clouds of the real-time static object boundary plane in sequence And the corresponding observation direction vector , and calculate the angle between the current boundary direction vector and the observation direction vector , forming a one-dimensional vector As a real-time static object boundary environment representation, it can realize the extraction of static object boundary direction trend in the real-time perception scenario on the vehicle side; Among them, the current direction vector is calculated by the following method: ； in, Indicates the number of columns of the vector, i.e. the number of plane splitting rays, and Respectively represent the position coordinates of the nth and n-1th real-time point clouds, Indicates the direction vector from the n-1th real-time static object boundary equivalent input point cloud to the nth real-time static object boundary equivalent input point cloud, x n and n For vector The vector coordinates in the XOY plane of the vehicle coordinate system centered on the current vehicle observation point are also the position coordinates of the equivalent input point cloud of the static object boundary at the nth time.

5. An automatic driving global position estimation system based on static object association of standard definition electronic map, characterized in that: include: Data acquisition module, used to obtain standard-definition electronic maps in the scenario of autonomous driving city operation; The parsing and processing module is used to parse and preprocess the standard definition electronic map to extract a set of candidate vehicle position nodes during automatic driving, including: obtaining road network data corresponding to the standard definition electronic map, parsing the connection relationship of the nodes in the standard definition electronic map; filtering the nodes by selecting road attributes, and screening out roads that are passable by automatic driving vehicles as the vehicle drivable domain road set P', the vehicle drivable domain road set P' including a number of original road network node coordinates p' (x, y); for two adjacent road nodes p1' (x1, y1) and p2' (x2, y2) belonging to the same road, respectively calculating the distance between the two road nodes The plane geometric distance and the horizontal and vertical coordinate difference between the two adjacent road nodes are calculated, and the number of nodes to be interpolated is determined according to the preset node interval; the coordinate increment p3′ (X increment, Y increment) of each point to be interpolated is calculated according to the difference X between the horizontal coordinates of the two adjacent road nodes, the difference Y between the vertical coordinates and the number of nodes to be interpolated; the points to be interpolated are distributed at equal distances in the horizontal and vertical directions between the two road nodes according to the coordinate difference between the two road nodes, the number of nodes to be interpolated and the corresponding coordinate increments; each original road network node in the set P' of all drivable domain roads in the standard definition electronic map is traversed and interpolated according to the above method to form a set of candidate vehicle position nodes in the electronic map. ; Wherein, X difference = x2-x1, Y difference = y2-y1, X increment = X difference / number of nodes to be interpolated, Y increment = Y difference / number of nodes to be interpolated, N represents the number of road nodes after interpolation in the electronic map; p1, p2, ... pN represent the plane node coordinates corresponding to the 1st, 2nd, ... Nth road nodes; An electronic map extraction module is used to determine and screen the static feature boundaries within a preset range for the vehicle position candidate node set, and after screening, extract the city structured static feature description information on the standard definition electronic map to construct a priori static feature environment description database; A real-time perception and extraction module is used to extract the description of urban structured static objects through real-time perception of vehicles and input three-dimensional point cloud data in real time, so as to obtain real-time static object environment description data; The position estimation module is used to associate the static object data in the standard definition electronic map with the real-time static object environment description data perceived by the vehicle in real time, and obtain the position candidate result with the highest similarity to the real-time perception of the vehicle to achieve the global position estimation of the autonomous driving vehicle, including: By traversing each element in the prior static object environment description database, the most similar result to the real-time perception is queried in the prior static object environment description database, and the similarity calculation is implemented using the n-norm method as follows: , Where D(i) represents the similarity between the input static object environment description vector perceived in real time and the i-th candidate in the prior static object environment description database. Represents the reference vector corresponding to the real-time perceived static feature description, represents the i-th static object description vector corresponding to each vehicle global position candidate in the database, and N is the total number of static object environment description vectors in the prior static object environment description database; The calculated similarity results are compared to select the candidate result with the highest similarity to the input perception vector, and the vehicle's global position estimate is obtained by accessing the candidate coordinates of the vehicle's global position corresponding to the optimal candidate result; The parsing and preprocessing of the standard definition electronic map includes: Obtaining the road network data corresponding to the standard definition electronic map, parsing the connection relationship of the nodes in the standard definition electronic map, and extracting all the static feature boundary information nodes in the parsed standard definition electronic map; Filter nodes by selecting boundary attributes to select target static objects with boundary, wherein the boundary attributes include buildings, fences, barriers and road-related boundary objects; Divide according to the same attributes of the boundary and construct a static object set composed of static ground feature boundary points ; Where K is the number of static feature boundaries contained in the standard definition electronic map, and the element B in the set B is j Represents the set of nodes belonging to static objects corresponding to the same attribute, j=1, 2, ..., K; ; , ,... Represents the two-dimensional coordinates of the 1st, 2nd, ...nth nodes in the static feature in the electronic map.

6. An electronic device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the method for global position estimation of autonomous driving based on static object association of standard definition electronic map as described in any one of claims 1 to 4.

7. A readable storage medium, characterized in that: The readable storage medium stores a processing program, and when the processing program is executed by the processor, the method for estimating the global position of an autonomous driving based on the association of static objects in a standard-definition electronic map is implemented as described in any one of claims 1 to 4.

Images