Examination method and equipment for highway ADAS map

Abstract

The invention provides a highway ADAS map inspection method and device, and belongs to the technical field of automatic driving. Iteration is carried out by using a simulated annealing algorithm based on national highway network data, and an optimal solution obtained by iteration is used as a navigation path; according to the method, a plurality of navigation paths which more accurately cover all expressways throughout the country and are repeated as little as possible are generated, the ADAS map generated on the basis of the navigation paths is more accurate, and map data repetition is avoided, so that the workload is reduced, the inspection efficiency is improved and the inspection cost is reduced in the process of finally performing automatic inspection on the ADAS map.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method and device for checking a highway ADAS map. Background Art

[0002] ADAS (Advanced Driver Assistance Systems) map is a kind of map data designed for advanced driver assistance systems. ADAS map is a kind of map data between vehicle map and high-precision map, and its accuracy is usually within 1 to 5 meters. It is mainly used to provide environmental information required for vehicle perception and driving decision-making to improve driving safety and driving experience. ADAS map not only contains traditional road location information, but also adds data information of the road itself such as slope, curvature, heading angle, as well as road-related data information such as details of highway exits, road-related points of interest, road speed limit information, and traffic sign information.

[0003] Currently, in the field of autonomous driving, the data quality of ADAS maps is closely related to the assisted driving function.

[0004] Due to the large amount of national map data and diverse map attributes, it is easy to have duplicate paths when planning routes. ADAS maps are related to the location points in the path in real time, resulting in a large and complicated workload for traditional ADAS map inspection methods, low inspection efficiency and high cost. Summary of the invention

[0005] In view of this, the present application provides a method for checking a highway ADAS map, which can reduce the workload of automatic checking of the ADAS map. The method includes:

[0006] Get national highway network data.

[0007] Based on the national highway network data, the greedy algorithm and the simulated annealing algorithm are used to generate an optimized highway path. The optimized highway path includes multiple navigation paths. The multiple navigation paths correspond to the optimal solutions obtained by iterating the simulated annealing algorithm, and the union of the multiple navigation paths covers all highways in the national highway network data.

[0008] Generate ADAS maps based on optimized highway routes.

[0009] The ADAS map is staticized to obtain a static map.

[0010] Automated inspection of static maps.

[0011] Optionally, based on national highway network data, the greedy algorithm and simulated annealing algorithm are used to generate an optimized highway path including:

[0012] Convert national highway network data into a graph structure.

[0013] Generate a minimum spanning tree of a graph structure based on the graph structure.

[0014] The greedy algorithm is used to supplement the minimum spanning tree and obtain the covering data.

[0015] The simulated annealing algorithm is used to iterate the coverage data to obtain the optimal solution, and multiple navigation paths corresponding to the optimal solution are used as optimized high-speed paths.

[0016] Optionally, generating an ADAS map based on the optimized highway path includes:

[0017] A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

[0018] Optionally, the ADAS map is staticized to obtain a static map including:

[0019] The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain a static map.

[0020] Optionally, automated checks on static maps include:

[0021] At least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check is performed on the static map.

[0022] The present application also provides a highway ADAS map inspection device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, an automated highway ADAS map inspection method is implemented, the method comprising:

[0023] Get national highway network data.

[0024] Based on the national highway network data, the greedy algorithm and the simulated annealing algorithm are used to generate an optimized highway path. The optimized highway path includes multiple navigation paths. The multiple navigation paths correspond to the optimal solutions obtained by iterating the simulated annealing algorithm, and the union of the multiple navigation paths covers all highways in the national highway network data.

[0025] Generate ADAS maps based on optimized highway routes.

[0026] The ADAS map is staticized to obtain a static map.

[0027] Automated inspection of static maps.

[0028] Optionally, based on national highway network data, the greedy algorithm and simulated annealing algorithm are used to generate an optimized highway path including:

[0029] Convert national highway network data into a graph structure.

[0030] Generate a minimum spanning tree of a graph structure based on the graph structure.

[0031] The greedy algorithm is used to supplement the minimum spanning tree and obtain the covering data.

[0032] The simulated annealing algorithm is used to iterate the coverage data to obtain the optimal solution, and multiple navigation paths corresponding to the optimal solution are used as optimized high-speed paths.

[0033] Optionally, generating an ADAS map based on the optimized highway path includes:

[0034] A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

[0035] Optionally, the ADAS map is staticized to obtain a static map including:

[0036] The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain a static map.

[0037] Optionally, automated checks on static maps include:

[0038] At least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check is performed on the static map.

[0039] The highway ADAS map inspection method provided in the present application is adopted, and a simulated annealing algorithm is used to iterate based on the national highway network data to determine the optimal solution, and generate multiple navigation paths that are more accurate and cover all highways in the country with as few path duplications as possible. On this basis, the ADAS map generated based on the navigation path is also more accurate and avoids duplication of map data, thereby reducing the workload, improving inspection efficiency and reducing inspection costs in the final process of automated inspection of the ADAS map. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1A flowchart of a method for checking a highway ADAS map provided in an embodiment of the present application;

[0042] Figure 2 Another flow chart of a method for checking a highway ADAS map provided in an embodiment of the present application;

[0043] Figure 3 An architectural diagram of a highway ADAS map inspection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] Currently in the field of autonomous driving, the quality of ADAS map data is closely related to the assisted driving function. In the current industry, the traditional map data inspection method mainly relies on operators to perform manual verification. Due to the large amount of national map data and the diverse map attributes, this method is labor-intensive and complicated to implement, with low update efficiency and high labor costs. Moreover, ADAS map data is sent dynamically based on the current location, and its inspection is also different from the inspection of traditional static maps. Therefore, it is urgent to establish an efficient and reliable ADAS map data automatic inspection method to ensure the quality of ADAS map data and reduce the workload of automated inspection of ADAS maps.

[0046] The present application embodiment provides a method for checking a highway ADAS map, such as Figure 1 As shown, the method includes steps S101, S102, S103, S104 and S105:

[0047] In step S101, national highway network data is obtained.

[0048] In step S102, based on the national highway network data, an optimized highway route is generated using a greedy algorithm and a simulated annealing algorithm.

[0049] The optimized expressway path includes multiple navigation paths, the multiple navigation paths correspond to the optimal solutions obtained by iteration according to the simulated annealing algorithm, and the union of the multiple navigation paths covers all expressways in the national expressway network data.

[0050] In step S103 , an ADAS map is generated based on the optimized highway route.

[0051] In step S104, the ADAS map is staticized to obtain a static map.

[0052] In step S105 , an automated check is performed on the static map.

[0053] In some optional embodiments, based on national highway network data, generating an optimized highway path using a greedy algorithm and a simulated annealing algorithm includes:

[0054] Convert national highway network data into a graph structure.

[0055] Generate a minimum spanning tree of a graph structure based on the graph structure.

[0056] The greedy algorithm is used to supplement the minimum spanning tree and obtain the covering data.

[0057] The simulated annealing algorithm is used to iterate the coverage data to obtain the optimal solution, and multiple navigation paths corresponding to the optimal solution are used as optimized high-speed paths.

[0058] In some optional embodiments, generating an ADAS map based on the optimized highway path includes:

[0059] A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

[0060] In some optional embodiments, the ADAS map is staticized to obtain a static map including:

[0061] The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain a static map.

[0062] In some optional embodiments, the automated inspection of the static map includes:

[0063] At least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check is performed on the static map.

[0064] By adopting the inspection method of highway ADAS map provided by the present application, a simulated annealing algorithm is used to iterate based on the national highway network data, and the optimal solution obtained by the iteration is used as the navigation path to generate more accurate multiple navigation paths that cover all highways in the country with as few path duplications as possible. On this basis, the ADAS map generated based on the navigation path is also more accurate and avoids duplication of map data, thereby reducing the workload, improving inspection efficiency and reducing inspection costs in the final process of automated inspection of the ADAS map.

[0065] The present application also provides a method for checking a highway ADAS map. Figure 2As shown, the method includes steps S201, S202, S203, S204, S205, S206, S207 and S208:

[0066] In step S201, national highway network data is obtained.

[0067] It is understandable that the national highway network data can be pre-stored or obtained from a server.

[0068] In step S202, the national highway network data is converted into a graph structure.

[0069] It can be understood that the national highway network is first represented by a graph structure, where nodes represent intersections of highways (such as exits, entrances, toll stations, etc., and the entrances and exits can be found at the connection between the ramp and the highway), edges represent highway sections, and distance is used to represent the weight of the edge.

[0070] In step S203, a minimum spanning tree of the graph structure is generated based on the graph structure.

[0071] It is understandable that after using the graph structure to represent the national highway network, the minimum spanning tree of the graph structure is generated. The minimum spanning tree is a tree that covers all nodes in the graph structure and has the smallest total weight of the edges. In this way, a planned path that covers all entrances and exits with the shortest distance can be obtained. It can be used as a basic path for traveling across the national highway because it does not have repeated sections.

[0072] In step S204, the minimum spanning tree is supplemented with a greedy algorithm to obtain coverage data.

[0073] It is understandable that after generating the minimum spanning tree of the graph structure of the national highway network, some sections may not be covered because the minimum spanning tree has no repeated edges. In order to cover all highway sections, a greedy algorithm can be used to supplement the coverage. Specifically, starting from any node in the graph, select the shortest section among all uncovered sections of the current node, then add this section to the path and mark it as covered. Repeat the above process until all sections are covered, and finally process the remaining isolated sections separately. In this way, a local optimal solution covering all edges can be temporarily obtained.

[0074] In step S205, the coverage data is iterated using a simulated annealing algorithm to obtain an optimal solution, and multiple navigation paths corresponding to the optimal solution are used as optimized high-speed paths.

[0075] The optimized expressway path includes multiple navigation paths, the multiple navigation paths correspond to the optimal solutions obtained by iteration according to the simulated annealing algorithm, and the union of the multiple navigation paths covers all expressways in the national expressway network data.

[0076] It is understandable that the simulated annealing algorithm is finally chosen to further optimize the path. It is an algorithm for global problem optimization. Its goal is to find the global optimal solution (i.e., the optimal solution in step S205) starting from a certain initial solution. The core of its algorithm is the "temperature" parameter T, which gradually decreases over time. When the temperature is high, the algorithm can accept poor solutions, which increases the randomness and diversity of the search. When the temperature gradually decreases, the algorithm is more inclined to choose a better solution. The acceptance probability formula is shown in formula (1):

[0077]

[0078] Where P is the probability, e represents the base of the natural logarithm, i.e., the natural constant, ΔE is the difference between the objective function of the current solution and the new solution, and T is the temperature. During the search process, the algorithm starts from an initial point, then adjusts according to the temperature and randomly selects a neighborhood solution. If the solution is better than the current solution, it is directly accepted; if it is worse, it is decided whether to accept it according to the probability P. As the temperature gradually decreases, the algorithm's exploration pace gradually decreases, and finally a detailed search is performed in a local area, gradually converging to the global optimal solution.

[0079] Specifically, we first set an initial temperature T0 and a cooling coefficient c (0.8-0.99). After the greedy algorithm obtains an initial path, we continuously try to optimize the current solution locally (such as node exchange, reversing part of the path, reconnecting the path, etc.). The objective function is the weight and E of the solution. Based on the calculated probability value P, we determine whether to accept the new solution. If accepted, we update the path. Regardless of whether it is accepted or not, after each iteration, the temperature T is multiplied by the cooling coefficient c, and the search stops when the convergence condition is met (the temperature drops to a certain threshold or the maximum number of iterations is reached).

[0080] In this way, a set of paths that are close to the global optimal can be automatically generated and used as the optimized highway path, which can effectively cover the national highways and minimize repeated sections.

[0081] More specifically, the solution obtained by the greedy algorithm is the input of the simulated annealing algorithm, which is obtained through DFS (depth-first traversal). The solution obtained by the greedy algorithm is an array containing different paths, and each array is a collection of edges. The edges of the directed graph generated by the greedy algorithm are weighted, and the weight value is the length of the corresponding edge. The weight sum of an initial solution can be obtained by calculation, and then a feedback function is added to the total length of the generated path and the number of special points (road merge, split point). That is, the closer the length is to the median and the closer the number of special points is to 2, the closer it can be considered that the path is to the desired path. The neighborhood solution is a new solution after local optimization, which is also a path. According to the probability P calculated by formula (1), the old solution obtained by the greedy algorithm is used as input. If the corresponding probability P is greater than a specified threshold, the old solution is used as the new solution, otherwise it is not used as the new solution.

[0082] In step S206 , an ADAS map is generated based on the optimized highway route.

[0083] In some optional embodiments, generating an ADAS map based on the optimized highway path includes:

[0084] A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

[0085] In step S207, the ADAS map is staticized to obtain a static map.

[0086] In some optional embodiments, the ADAS map is staticized to obtain a static map including:

[0087] The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain a static map.

[0088] It is understandable that the ADAS map can be converted through a pre-stored ADAS map data visualization method designed based on QGIS to obtain a converted traditional static map.

[0089] A QGIS-based ADAS map data visualization method specifically includes:

[0090] 1. Record the GPS coordinates of the vehicle’s driving trajectory and ADAS map data.

[0091] 2. Segment the GPS data, aggregate the GPS points in the same segment to form the coordinate points of the segment, use these coordinate points as line geometry in geojson, and use the map information in the segment as properties in geojson to generate the corresponding segment.geojson file.

[0092] 3. The stub is point data. At the segmentation of each segment, that is, the GPS point at the segmentation in the segment is the location point of the stub. The coordinate point is used as the point geometry in geojson, and the map information in the stub is used as the properties in geojson to generate the corresponding stub.geojson file.

[0093] 4. The short profile is point data. According to the offset in the ADAS map, the nearest GPS point in the corresponding segment is found as the location of the data. That is, the coordinate point is used as the point geometry in geojson, and the map information in short is used as the properties in geojson, thereby generating a short.geojson file.

[0094] 5. The long profile operation is similar to the shrot profile operation, and the corresponding long.geojson file is generated.

[0095] 6. QGIS loads the processed geojson file and satellite image to visualize the ADAS map data, that is, convert the ADAS map into a converted traditional static map.

[0096] In step S208, an automated check is performed on the static map.

[0097] In some optional embodiments, the automated inspection of the static map includes:

[0098] At least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check is performed on the static map.

[0099] It is understandable that in order to ensure the integrity and accuracy of the generated map data, some quantitative indicators and inspection items can be defined to conduct a round of automated inspection of the data in advance, and then manual verification can be performed on areas where the indicators are abnormal. This can greatly reduce the amount of manual verification content and thus improve efficiency.

[0100] Automated inspection includes but is not limited to the following aspects:

[0101] 1. Data integrity check, including:

[0102] Missing data detection: Use graph algorithms to traverse the map network to check whether there are isolated nodes or unconnected sections, and verify whether there is missing key data in the map, such as roads, intersections, traffic signs, extended lanes, etc.

[0103] Redundant data detection: Check whether there are duplicate road segments, nodes or other redundant data.

[0104] 2. Geographic consistency check, including:

[0105] Topological consistency: Ensure that the topological structure of the map data is correct. For example, check whether all road segments are correctly connected, and whether the angles, distribution, and locations of intersections are reasonable.

[0106] Geometric consistency: Verify the accuracy of geometric shapes, such as checking whether the geometry of roads and lanes meets expectations through geometric analysis algorithms (such as Bezier curves), or restoring geometry through lane curvature.

[0107] 3. Check the accuracy of attributes, including:

[0108] Rule-based range check: Checks whether the attributes of map data are reasonable based on a predefined set of rules (such as speed limit range, road segment type, etc.). For example, verify whether the road type (highway, urban road, etc.), speed limit information, and road direction (one-way, two-way) are accurate.

[0109] Based on true value comparison check: If there is true value data from other map sources, the attributes at the same location can be compared to determine whether the current map data attributes are correct. For example, verify traffic signs (location, type), special roads (location, type, number of lanes, lane numbers), etc.

[0110] 4. Version difference check, including comparison of map data of the same path in different versions to ensure that the differences come from version changes rather than data changes.

[0111] 5. Simulation and scenario coverage check, including checking the performance of the autonomous driving system in various map scenarios through simulation tests, such as the performance of the autonomous driving system under different road types, weather conditions, and traffic density. The robustness and reliability of the map data can be verified by outputting the coverage and accuracy of the map data in different scenarios.

[0112] By adopting the inspection method of highway ADAS map provided by the present application, a simulated annealing algorithm is used to iterate based on the national highway network data, and the optimal solution obtained by the iteration is used as the navigation path to generate more accurate multiple navigation paths that cover all highways in the country with as few path duplications as possible. On this basis, the ADAS map generated based on the navigation path is also more accurate and avoids duplication of map data, thereby reducing the workload, improving inspection efficiency and reducing inspection costs in the final process of automated inspection of the ADAS map.

[0113] like Figure 3 As shown, the embodiment of the present application further provides a highway ADAS map inspection device 30, including a processor 301, a memory 302, and a computer program stored in the memory 302 and executable on the processor 301. When the computer program is executed by the processor 301, a highway ADAS map inspection method is implemented, and the method includes:

[0114] Get national highway network data.

[0115] Based on the national highway network data, the greedy algorithm and simulated annealing algorithm are used to generate optimized highway routes.

[0116] The optimized expressway path includes multiple navigation paths, the multiple navigation paths correspond to the optimal solutions obtained by iteration according to the simulated annealing algorithm, and the union of the multiple navigation paths covers all expressways in the national expressway network data.

[0117] Generate ADAS maps based on optimized highway routes.

[0118] The ADAS map is staticized to obtain a static map.

[0119] Automated inspection of static maps.

[0120] In some optional embodiments, based on national highway network data, generating an optimized highway path using a greedy algorithm and a simulated annealing algorithm includes:

[0121] Convert national highway network data into a graph structure.

[0122] Generate a minimum spanning tree of a graph structure based on the graph structure.

[0123] The greedy algorithm is used to supplement the minimum spanning tree and obtain the covering data.

[0124] The simulated annealing algorithm is used to iterate the coverage data to obtain the optimal solution, and multiple navigation paths corresponding to the optimal solution are used as optimized high-speed paths.

[0125] In some optional embodiments, generating an ADAS map based on the optimized highway path includes:

[0126] A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

[0127] In some optional embodiments, the ADAS map is staticized to obtain a static map including:

[0128] The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain a static map.

[0129] In some optional embodiments, the automated inspection of the static map includes:

[0130] At least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check is performed on the static map.

[0131] The inspection equipment for the highway ADAS map provided by the present application uses a simulated annealing algorithm to iterate based on the national highway network data, and uses the optimal solution obtained by the iteration as the navigation path to generate more accurate multiple navigation paths that cover all highways in the country with as few path duplications as possible. On this basis, the ADAS map generated based on the navigation path is also more accurate and avoids duplication of map data, thereby reducing the workload, improving inspection efficiency, and reducing inspection costs in the final process of automated inspection of the ADAS map.

[0132] The embodiment of the present application also provides a computer-readable storage medium, such as a memory including a program code, and the program code can be executed by a processor of a highway ADAS map inspection device to complete the highway ADAS map inspection method in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device.

[0133] A person skilled in the art will appreciate that all or part of the steps for implementing the above embodiments may be accomplished by hardware or by hardware associated with a program code, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0134] In the present application, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0135] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only.

[0136] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

[0137] The above description is only for the purpose of facilitating those skilled in the art to understand the technical solution of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for checking a highway ADAS map, characterized in that: The method comprises: Obtain national highway network data; Based on the national highway network data, an optimized highway path is generated using a greedy algorithm and a simulated annealing algorithm, wherein the optimized highway path includes a plurality of navigation paths, the plurality of navigation paths correspond to optimal solutions obtained by iteration according to the simulated annealing algorithm, and a union of the plurality of navigation paths covers all highways in the national highway network data; generating an ADAS map based on the optimized highway path; Stabilizing the ADAS map to obtain a static map; An automated check is performed on the static map.

2. The method for checking a highway ADAS map according to claim 1, characterized in that: The method of generating an optimized highway path based on the national highway network data by using a greedy algorithm and a simulated annealing algorithm includes: Converting the national highway network data into a graph structure; Generate a minimum spanning tree of the graph structure based on the graph structure; Using a greedy algorithm to perform supplementary coverage on the minimum spanning tree to obtain coverage data; The coverage data is iterated using a simulated annealing algorithm to obtain the optimal solution, and the multiple navigation paths corresponding to the optimal solution are used as the optimized high-speed paths.

3. The method for checking a highway ADAS map according to claim 1, characterized in that: Generating an ADAS map based on the optimized highway path includes: A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized high-speed path.

4. The method for checking a highway ADAS map according to claim 1, characterized in that: The step of staticizing the ADAS map to obtain a static map comprises: The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain the static map.

5. The method for checking a highway ADAS map according to claim 1, characterized in that: The automatic inspection of the static map comprises: The static map is subjected to at least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check.

6. A highway ADAS map inspection device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by the processor, a method for automatically checking a highway ADAS map is implemented, wherein the method comprises: Obtain national highway network data; Based on the national highway network data, an optimized highway path is generated using a greedy algorithm and a simulated annealing algorithm, wherein the optimized highway path includes a plurality of navigation paths, the plurality of navigation paths correspond to optimal solutions obtained by iteration according to the simulated annealing algorithm, and a union of the plurality of navigation paths covers all highways in the national highway network data; generating an ADAS map based on the optimized highway path; Stabilizing the ADAS map to obtain a static map; An automated check is performed on the static map.

7. The highway ADAS map inspection device according to claim 6, characterized in that: The method of generating an optimized highway path based on the national highway network data by using a greedy algorithm and a simulated annealing algorithm includes: Converting the national highway network data into a graph structure; Generate a minimum spanning tree of the graph structure based on the graph structure; Using a greedy algorithm to perform supplementary coverage on the minimum spanning tree to obtain coverage data; The coverage data is iterated using a simulated annealing algorithm to obtain the optimal solution, and the multiple navigation paths corresponding to the optimal solution are used as the optimized high-speed paths.

8. The highway ADAS map inspection device according to claim 6, characterized in that: Generating an ADAS map based on the optimized highway path includes: A corresponding ADAS map is generated according to at least one GPS point corresponding to the optimized highway path.

9. The highway ADAS map inspection device according to claim 6, characterized in that: The step of staticizing the ADAS map to obtain a static map comprises: The ADAS map is staticized using a pre-stored ADAS map data visualization algorithm to obtain the static map.

10. The highway ADAS map inspection device according to claim 6, characterized in that: The automatic inspection of the static map comprises: The static map is subjected to at least one of a data integrity check, a geographic consistency check, an attribute accuracy check, a version difference check, and a simulation and scene coverage check.

Images