Methods, devices, equipment, media, and vehicles for high-precision map connectivity verification

Abstract

This invention discloses a method, apparatus, device, medium, and vehicle for verifying the connectivity of high-precision maps. The method includes: extracting road centerlines from the high-precision map whose connectivity needs to be verified, and matching the extracted road centerlines with corresponding skeleton lines; in the matching results, if there are road segments in the skeleton lines that are not matched with the extracted road centerlines, it is determined whether the unmatched road segments are road connecting segments, wherein both ends of a road connecting segment have other road segments topologically connected to the road connecting segment; if the unmatched road segments are not road connecting segments, it is determined that the connectivity of the high-precision map to be verified is not missing; if the unmatched road segments are road connecting segments, it is determined that the connectivity of the high-precision map to be verified is missing. By adopting the above technical solution, the efficiency of high-precision map connectivity verification is improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of autonomous driving technology, and more specifically, to a method, apparatus, device, medium, and vehicle for high-precision map connectivity verification. Background Technology

[0002] In the process of creating high-precision maps, problems such as missing road segments at the end of roads or interruptions in the connectivity of roads that should be continuous are often caused by issues in the production process.

[0003] In related technologies, maps are built at a tile level. After each tile's high-precision map is completed, map connectivity is verified. The specific verification method involves overlaying the base map onto the newly created high-precision map. This allows manual verification of missing continuous road segments in the new high-precision map based on the color differences between the base map and the high-precision map. This manual verification method is time-consuming and labor-intensive, resulting in relatively low efficiency. Summary of the Invention

[0004] This invention provides a method, apparatus, device, medium, and vehicle for verifying the connectivity of high-precision maps, thereby improving the efficiency of high-precision map connectivity verification.

[0005] The specific technical solution is as follows:

[0006] In a first aspect, embodiments of the present invention provide a connectivity verification method for a high-precision map, comprising:

[0007] Extract the road centerlines from the high-precision map whose connectivity needs to be verified, and match the extracted road centerlines with the corresponding skeleton lines. The matching content includes road label information, road location information, and road angle information. The skeleton lines are traditional maps used to create high-precision maps, which are produced by open-source map generation.

[0008] If all matching items of the extracted road centerline and the corresponding skeleton line are successful, it is determined that the connectivity of the high-precision map to be verified is not missing.

[0009] In the matching results, if there are road segments in the skeleton line that are not matched with the extracted road centerline, it is determined whether the unmatched road segments are road connecting segments. Among them, there are other road segments that are topologically connected to the road connecting segment at both ends.

[0010] If the unmatched road segment is not a road connection segment, then it is determined that the connectivity of the high-precision map to be verified is not missing.

[0011] If the unmatched road segment is a road connection segment, it is determined that the connectivity of the high-precision map to be verified is missing.

[0012] As shown in the above scheme, by extracting the road centerlines from the high-precision map whose connectivity needs to be verified and matching them with the corresponding skeleton lines, it can be determined whether the connectivity of the high-precision map to be verified is missing based on the matching results. If there are road segments in the matching results that do not match the high-precision map, and these road segments are road connections, then it can be determined that the connectivity of the high-precision map is missing. If the road segments that do not match the high-precision map are not road connections, or if the high-precision map matches the skeleton lines successfully, then it can be determined that the connectivity of the high-precision map to be verified is not missing. Compared with the manual verification of high-precision map connectivity provided by related technologies, this embodiment effectively improves the efficiency and accuracy of map connectivity verification by adopting the above-mentioned automated connectivity verification method.

[0013] Optionally, the method provided in this embodiment of the invention further includes:

[0014] The high-precision map with missing connectivity is rebuilt, and the rebuilt high-precision map is used as the new high-precision map to be verified for connectivity. Then, the road center lines of the new high-precision map to be verified are extracted, and the matching operation between the road center lines and the corresponding skeleton lines is performed until it is determined that there are no missing connectivity in the high-precision map to be verified.

[0015] Optionally, the extracted road centerline is matched with the corresponding skeleton line, including:

[0016] Based on the road width of each segment in the skeleton line, determine the corresponding road range of each segment in the skeleton line; then match the matching content corresponding to each high-precision segment in the extracted road centerline with the matching content corresponding to the road range of each segment in the skeleton line.

[0017] As can be seen from the above technical solution, by matching the road center line segment of the high-precision map with the road range of each road segment in the skeleton line, the mismatch or omission caused by the road center line of the skeleton line not being in the middle of the road is avoided, thus improving the accuracy of the matching results.

[0018] Optionally, the high-precision map is constructed at the cell level, with overlapping areas between adjacent cells. Each cell is divided according to different high-precision road segments in the high-precision map.

[0019] Correspondingly, the connectivity of high-precision maps is verified at the cell level;

[0020] Accordingly, after connectivity verification has been completed for all high-precision maps in each cell, the method provided in this embodiment of the invention further includes:

[0021] The cells are merged according to overlapping road segments to obtain a complete high-precision map.

[0022] Optionally, if there are no missing connections in the high-precision map of each cell, the method provided in this embodiment of the invention further includes:

[0023] For the first road segment in the successfully matched road centerline and the second road segment in the corresponding matched skeleton line, the straight road segments with topological connections in the successfully matched road centerline segments are merged into the same first road segment to be matched, and the straight road segments with topological connections in the corresponding matched skeleton line segments are merged into the same second road segment to be matched.

[0024] Each first road segment to be matched is matched with each second road segment to be matched to obtain the first and second road segments to be matched successfully with the preset matching conditions. The preset matching conditions include distance information, angle information, and observation length coverage. The observation length coverage includes: the ratio of the length of the second road segment observed by the first road segment to the length of the first road segment, and the ratio of the length of the first road segment observed by the second road segment to the length of the second road segment.

[0025] Replace the accuracy parameter value of the second road segment that was successfully matched with the accuracy parameter value of the first road segment that was successfully matched with it. The accuracy parameters include longitude information, latitude information and elevation information.

[0026] As can be seen from the above technical solution, by correcting the accuracy of the skeleton lines at the corresponding locations using the accuracy of the high-precision map, a more accurate skeleton line can be obtained for use in the creation of subsequent high-precision maps.

[0027] Optionally, each first road segment to be matched is matched with each second road segment to be matched to obtain the first and second road segments to be matched that achieve the optimal matching conditions, including:

[0028] First implementation method:

[0029] Based on the Hungarian algorithm, each first road segment to be matched is matched with each second road segment to be matched, so as to obtain the first and second road segments to be matched that achieve the optimal matching conditions.

[0030] The second implementation method:

[0031] For any first road segment to be matched, the matching score of each second road segment to be matched with the first road segment is determined according to the preset matching conditions. The second road segment with the largest score is selected from the matching scores and is used as the second road segment that has been successfully matched with the first road segment. This process continues until all first road segments have been traversed, and the second road segments that have been successfully matched with each first road segment are obtained.

[0032] Secondly, embodiments of the present invention also provide a connectivity verification device for a high-precision map, comprising:

[0033] The matching module is configured to extract the road centerline of the high-precision map to be verified for connectivity, and match the road centerline with the skeleton line at the corresponding location. The matching content includes road label information, road location information and road angle information. The skeleton line is a traditional map obtained from open source map production for creating the high-precision map.

[0034] The connectivity determination module is configured to determine that the connectivity of the high-precision map to be verified is not missing if all matching contents of the road centerline and the corresponding skeleton line are successful.

[0035] In the matching results, if there are road segments in the skeleton line that are not matched with the road centerline, it is determined whether the unmatched road segments are road connecting segments, wherein both ends of the road connecting segment have other road segments that are topologically connected to the road connecting segment.

[0036] If the unmatched road segment is not a road connection segment, then it is determined that the connectivity of the high-precision map to be verified is not missing.

[0037] If the unmatched road segment is a road connection segment, it is determined that the connectivity of the high-precision map to be verified is missing.

[0038] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0039] The re-verification module is configured to reconstruct the high-precision map with missing connectivity, and use the reconstructed high-precision map as the new high-precision map to be verified for connectivity. It then returns to perform the operation of extracting the road centerlines from the new high-precision map to be verified for connectivity, and matching the road centerlines with the corresponding skeleton lines, until it is determined that the connectivity of the high-precision map to be verified is not missing.

[0040] Optionally, the matching module is specifically configured as follows:

[0041] Extract the road centerlines from the high-precision map whose connectivity needs to be verified;

[0042] Determine the corresponding road range for each segment in the skeleton line based on the road width of each segment.

[0043] Match the matching contents corresponding to each high-precision road segment in the road centerline with the matching contents corresponding to each road range in the skeleton line.

[0044] Optionally, the high-precision map is constructed at the cell level, with overlapping areas between adjacent cells. Each cell is divided according to different high-precision road segments in the high-precision map.

[0045] Correspondingly, the connectivity of high-precision maps is verified at the cell level;

[0046] Accordingly, after connectivity verification has been completed for all high-precision maps in each cell, the apparatus provided in this embodiment of the invention further includes:

[0047] The merging module is configured to merge cells according to overlapping road segments to obtain a complete high-precision map.

[0048] Optionally, if there are no missing connections in the high-precision map of each cell, the apparatus provided in this embodiment of the invention further includes:

[0049] The matching object segmentation module is configured to, for the first road segment in the successfully matched road centerline and the second road segment in the corresponding matched skeleton line, merge the straight road segments with topological connections in the successfully matched road centerline segment into the same first road segment to be matched, and merge the straight road segments with topological connections in the corresponding matched skeleton line segment into the same second road segment to be matched.

[0050] The road matching module is configured to match each first road segment to be matched with each second road segment to be matched, to obtain the first and second road segments to be matched successfully with the preset matching conditions. The preset matching conditions include distance information, angle information, and observation length coverage. The observation length coverage includes the ratio of the length of the second road segment observed by the first road segment to the length of the first road segment, and the ratio of the length of the first road segment observed by the second road segment to the length of the second road segment.

[0051] The skeleton line accuracy correction module is configured to replace the accuracy parameter value of the second matching road segment that has been successfully matched with the accuracy parameter value of the first matching road segment that has been successfully matched with it. The accuracy parameters include: longitude information, latitude information and elevation information.

[0052] Optionally, the road matching module may include:

[0053] The first matching unit is configured to match each first road segment to be matched with each second road segment to be matched based on the Hungarian algorithm, so as to obtain the first and second road segments to be matched that achieve the optimal matching conditions.

[0054] or,

[0055] The second matching unit is configured to, for any first road segment to be matched, determine the matching score of each second road segment to be matched with the first road segment according to the preset matching conditions, and select the second road segment corresponding to the largest score from the matching scores as the second road segment to be matched successfully matched with the first road segment, until all first road segments to be matched have been traversed, and the second road segment to be matched successfully matched with each first road segment is obtained.

[0056] Thirdly, embodiments of the present invention provide an electronic device, the electronic device comprising:

[0057] One or more processors;

[0058] Storage device for storing one or more programs.

[0059] When the one or more programs are executed by the one or more processors, the one or more processors implement the connectivity verification method for high-precision maps as provided in any embodiment of the present invention.

[0060] Fourthly, embodiments of the present invention provide a storage medium storing a computer program thereon, characterized in that, when the program is executed by a processor, it implements the connectivity verification method for high-precision maps as provided in any embodiment of the present invention.

[0061] Fifthly, embodiments of the present invention provide a vehicle that includes a connectivity verification device for a high-precision map provided in any embodiment of the present invention, or an electronic device provided in any embodiment of the present invention.

[0062] In a sixth aspect, embodiments of the present invention provide a computer program, the computer program including program instructions, which, when executed by a computer, implement the connectivity verification method for high-precision maps as provided in any embodiment of the present invention. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1a This is a flowchart of a high-precision map connectivity verification provided in Embodiment 1 of the present invention;

[0065] Figure 1b This is a schematic diagram of a high-precision road provided in Embodiment 1 of the present invention;

[0066] Figure 2a This is a flowchart of a high-precision map connectivity verification method provided in Embodiment 2 of the present invention;

[0067] Figure 2b This is a schematic diagram of road matching when using the accuracy of a high-precision map to correct the accuracy of the skeleton line, as provided in Embodiment 2 of the present invention.

[0068] Figure 3 This is a structural block diagram of a high-precision map connectivity verification device provided in Embodiment 3 of the present invention;

[0069] Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention;

[0070] Figure 5 This is a schematic diagram of a vehicle provided in Embodiment 5 of the present invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0073] This invention discloses a method, apparatus, device, medium, and vehicle for high-precision map connectivity verification. These are described in detail below.

[0074] Example 1

[0075] Figure 1a This is a flowchart illustrating a high-precision map connectivity verification method provided in Embodiment 1 of the present invention. This method can be applied to high-precision map connectivity verification scenarios. The method provided in this embodiment can be executed by a high-precision map connectivity verification device, which can be implemented through software and / or hardware. Figure 1a As shown, the method provided in this embodiment specifically includes:

[0076] S110. Extract the road centerline from the high-precision map to be verified for connectivity, and match the road centerline with the corresponding skeleton line. The matching content includes road label information, road location information, and road angle information.

[0077] First, it should be noted that during the high-precision map creation process, due to production technology or workflow issues, it is unavoidable that some road segments will be missing. The absence of the start and end segments of a road in a high-precision map is acceptable; that is, if the start and end segments of a road are not mapped, the high-precision map can be considered to have intact connectivity. However, if road connecting segments are missing, it indicates a lack of connectivity in the high-precision map. In this case, both ends of a road connecting segment must be topologically connected to other road segments. For example, Figure 1b This is a high-precision road diagram provided in Embodiment 1 of the present invention, as shown below. Figure 1b As shown, the high-precision road includes high-precision segment 1, high-precision segment 2, and high-precision segment 3. High-precision segment 1 is the starting segment, high-precision segment 2 is topologically connected to both high-precision segment 1 and high-precision segment 3, forming a connecting segment, and high-precision segment 3 is the ending segment. If either high-precision segment 1 or high-precision segment 3 is missing, other segments can still be used to enter and exit the high-precision road. Therefore, even when high-precision segment 1 and high-precision segment 3 are missing, the high-precision road can still be considered a fully connected road. However, if high-precision segment 2 is missing, the connectivity between the normally connected high-precision segment 1 and high-precision segment 3 is broken, resulting in a lack of connectivity for the high-precision road.

[0078] In this embodiment, the skeleton line is a map route generated from an open-source map (a map drawn by a user based on handheld devices, aerial photographs, or other free content, or even solely based on local knowledge, and freely usable by other users). It represents roads in the map using thin curves that match the original shape's connectivity and topological structure, serving as a representation of spatial data of geographical features. The road network within the skeleton line is spatially continuous, with an accuracy typically at the meter level. The skeleton line can serve as the foundation for creating high-precision maps. Those skilled in the art will understand that after creation, the skeleton line must ensure complete connectivity before it can be used as the basis for producing a high-precision map; that is, the connectivity of all roads within the skeleton line used to create the high-precision map must be complete. Therefore, this embodiment can match a skeleton line with complete road connectivity with a high-precision map whose connectivity needs to be verified to determine whether any high-precision road segments affecting road connectivity are missing from the high-precision map. The matching content includes road label information, road location information, and road angle information.

[0079] Specifically, in this embodiment, when matching the road centerline of the high-precision map with the corresponding skeleton line, the road centerline segment of the high-precision map can be matched with each road centerline segment in the skeleton line. Alternatively, to avoid mismatches or omissions caused by the road centerline of the skeleton line not being in the middle of the road, this embodiment can also divide the corresponding road range for each road segment of the skeleton line according to the road width of each road segment before matching, and match the road centerline segment of the high-precision map with the road range of each road segment in the skeleton line, thereby improving the accuracy of the matching results.

[0080] In addition, during the matching process, the skeleton line can be divided into multiple road segments, and each skeleton line segment can be matched with the road centerline of the high-precision map to simplify the matching calculation.

[0081] S120. If all matching items of the high-precision map's road centerline and the corresponding skeleton line are successfully matched, then the connectivity of the high-precision map to be verified is determined to be complete. In the matching results, if there are road segments in the skeleton line that do not match the road centerline of the high-precision map, then it is determined whether the unmatched road segment is a road connection segment. If the unmatched road segment is not a road connection segment, then the connectivity of the high-precision map to be verified is determined to be complete; if the unmatched road segment is a road connection segment, then the connectivity of the high-precision map to be verified is determined to be complete.

[0082] Those skilled in the art will understand that, assuming no connectivity issues exist in the high-precision map, during the matching process, each high-precision road segment in the high-precision map should have a corresponding matching object, i.e., a skeleton line segment to be matched with it, and each high-precision road segment should be able to successfully match with the corresponding skeleton line segment. That is, for each high-precision road segment belonging to the road centerline of the high-precision map, if all matching content between each high-precision road segment and the corresponding skeleton line segment is successfully matched, it indicates that the road centerline of the high-precision map is successfully matched with the corresponding skeleton line, meaning that the connectivity of the high-precision map to be verified is not lacking. However, if there are unmatched skeleton line segments in the matching results, i.e., the skeleton line segment does not have a corresponding matching object, it can be determined whether there are other road segments with topological connections at both ends of the unmatched road segment. If other road segments exist at both ends, it indicates that the unmatched skeleton line segment is a road connecting segment, for example... Figure 1b The high-precision road segment 2 is shown. If one end of the unmatched road segment has another road segment, or neither end has another road segment connected to the topological relationship, then it means that the unmatched skeleton road segment is not a road connection segment. For example, Figure 1b High-precision road segments 1 and 3 shown are not road connection segments. Determining whether other road segments with topological connections exist between the two ends of unmatched roads can be achieved using a depth-first search algorithm, which is not limited in this embodiment. In this embodiment, if the unmatched skeleton road segment is a road connection segment, it is determined that the connectivity of the high-precision map to be verified is incomplete; if the unmatched skeleton road segment is not a road connection segment, it is determined that the connectivity of the high-precision map to be verified is complete.

[0083] Specifically, the matching results of each skeleton line segment in the road centerline of the high-precision map, i.e., the road numbers of the matched skeleton line segments, can be combined to determine which road segments in the skeleton line were not matched, i.e., there is no corresponding matching object. For example, if there are ten road segments to be matched in the skeleton line segment, and the road numbers of the successfully matched high-precision road segments are 1 to 10, then it can be said that the road connectivity of the high-precision map to be verified is complete. However, if the road numbers of the matched high-precision road segments are 2 to 8, that is, three road segments in the skeleton line were not matched, then if the depth search method determines that these three road segments are road connected segments, it means that the high-precision map is missing road connected segments, i.e., the road connectivity of the high-precision map to be verified is missing. If the depth search method determines that the three missing road segments in the high-precision map are not road connected segments, then the road connectivity of the high-precision map to be verified is not missing.

[0084] In this embodiment, by extracting the road centerlines of the high-precision map whose connectivity needs to be verified, and matching the road centerlines with the corresponding skeleton lines, it can be determined whether the connectivity of the high-precision map to be verified is missing based on the matching results. If there are road segments in the matching results that do not match the high-precision map, and these road segments are road connections, then it can be determined that the connectivity of the high-precision map is missing. If the road segments that do not match the high-precision map are not road connections, or if the high-precision map matches the skeleton lines successfully, then it can be determined that the connectivity of the high-precision map to be verified is not missing. Compared with the manual verification of high-precision map connectivity provided by related technologies, this embodiment effectively improves the efficiency and accuracy of map connectivity verification by adopting the above-described automated connectivity verification method.

[0085] Furthermore, for high-precision maps with missing connectivity obtained from verification, the high-precision map can be reconstructed, and the reconstructed high-precision map can be used as a new high-precision map to be verified for connectivity. Then, the operation of extracting the road centerline of the new high-precision map to be verified for connectivity is performed, and the matching operation of the road centerline with the corresponding skeleton line is performed until it is determined that the connectivity of the high-precision map to be verified does not have missing connectivity.

[0086] Example 2

[0087] Figure 2a This is a flowchart of a high-precision map connectivity verification process provided in Embodiment 2 of the present invention. In this embodiment, the high-precision map to be verified is divided according to the granularity of cells. Different cells are divided according to road segments in the high-precision map, and there are overlapping areas between adjacent cells. In this embodiment, the connectivity verification process of the high-precision map is also performed sequentially at the cell granularity. After the connectivity verification of each cell's high-precision map is successful, the high-precision map of the cell with no missing connectivity can be used to optimize the skeleton line accuracy at the corresponding location, and the cells can be merged according to overlapping road segments to obtain a complete high-precision map corresponding to a certain map. Figure 2a As shown, the method provided in this embodiment includes:

[0088] S210. Perform connectivity verification on each cell of the high-precision map to be verified in sequence. During the connectivity verification process of the high-precision map in each cell, extract the road centerline of the high-precision map to be verified and match the road centerline with the skeleton line at the corresponding position. The matching content includes road label information, road location information and road angle information.

[0089] S220. Determine whether all matching items of the road centerline and the corresponding skeleton line of the high-precision map are successfully matched. If yes, proceed to step S230; otherwise, proceed to step S240.

[0090] S230. Determine that the connectivity of the high-precision map to be verified is not missing, and continue to execute step S270.

[0091] S240. Determine whether there are any skeleton line segments in the matching results that are not matched with the road centerline of the high-precision map. If so, proceed to step S250; otherwise, proceed to step S230.

[0092] S250. Determine whether the unmatched road segment is a road connecting segment. If so, proceed to step S260; otherwise, proceed to step S230.

[0093] S260. It has been determined that the connectivity of the high-precision map to be verified is incomplete.

[0094] Furthermore, assuming that there are no missing connections in the high-precision maps of all cells, the high-precision maps of each cell can be merged according to overlapping road segments to obtain a complete high-precision map of a certain region.

[0095] S270. For the first road segment in the successfully matched road centerline and the second road segment in the corresponding matched skeleton line, the straight road segments with topological connections in the successfully matched road centerline segments are merged into the same first road segment to be matched, and the straight road segments with topological connections in the corresponding matched skeleton line segments are merged into the same second road segment to be matched.

[0096] In this embodiment, for road segments in the successfully matched high-precision map and the corresponding skeleton line segments, the successful matching result includes the road number of each high-precision road segment and the road number of the skeleton line segment that successfully matches each high-precision road segment. In this embodiment, when using the accuracy of the high-precision map without missing parts to correct the accuracy of the corresponding skeleton line, in order to further accurately obtain the skeleton line segment that should be corrected for the high-precision road segment based on the above successful matching, this embodiment merges the straight road segments with topological connections in the successfully matched road center line segment results into the same first road segment to be matched, and merges the straight road segments with topological connections in the corresponding matched skeleton line segments into the same second road segment to be matched. This setting simplifies the matching calculation and obtains a more accurate one-to-one matching result, thereby allowing the accuracy correction of the skeleton line segment to be performed using the one-to-one matching relationship between the high-precision road segment and the skeleton line segment.

[0097] For example, Figure 2bThis is a schematic diagram illustrating road matching when using the accuracy of a high-precision map to correct the accuracy of skeleton lines, as provided in Embodiment 2 of the present invention. Figure 2b As shown, the road center segment AB and road center line segment BC in the high-precision map are topologically connected straight road segments. These two can be merged to obtain the second road segment to be matched, s1. The high-precision road segment BD serves as another straight road segment to be matched. The skeleton line segments ab and bc are topologically connected straight road segments. These two can be merged to obtain the first road segment to be matched, s2. The skeleton line segment bd serves as another straight road segment to be matched. After the merging operation is completed, the first road segment s1 and the second road segment s2 can be matched.

[0098] In this embodiment, if the road segment to be matched is a Y-shaped road segment, the Y-shaped road segment can be divided into three straight road segments for matching.

[0099] S280. Match each first road segment to be matched with each second road segment to be matched to obtain the first road segment to be matched and the second road segment to be matched that achieves the optimal matching conditions. Replace the precision parameter value of the second road segment to be matched with the precision parameter value of the first road segment to be matched with it.

[0100] In this embodiment, the following two methods can be used to match each first road segment to be matched with each second road segment to be matched, so as to obtain the first road segment to be matched and the second road segment to be matched successfully so that the preset matching conditions are optimal.

[0101] The preset matching conditions include the distance information between the first road segment to be matched and the second road segment to be matched, the angle information between the first road segment to be matched and the second road segment to be matched, and the observation length coverage rate. The observation length coverage rate includes the ratio of the length of the second road segment to be matched observed by the first road segment to the length of the first road segment, and the ratio of the length of the first road segment to the length of the second road segment observed by the second road segment.

[0102] Specifically, the observation length coverage can be calculated using the following method:

[0103] like Figure 2b As shown, multiple first observation points can be inserted at set intervals on the first road segment s1 to be matched. The total number of first observation points can be used to represent the length of the first road segment to be matched. The normal to the first observation point is drawn, for example, as shown... Figure 2bThe total number of intersection points on the second road segment to be matched with the normal L in the first observation point is the length of the second-generation matching road segment that can be observed from the first observation point. In this embodiment, the ratio of the length of the second road segment to be matched observed from the first observation point to the length of the first observation point is the ratio of the number of points on the second road segment to be matched that intersect with the normal L of the first observation point to the total number of points on the first observation point. For example, ... Figure 2b As shown, the observation length coverage of the road segment s1 to be matched for the road segment s2 is the ratio of the total number of intersection points (17) of the road segment s2 to be matched with the normal of the road segment s1 to the total number of the first observation points (17) of the road segment s1 to be matched, i.e., 100%.

[0104] Similarly, multiple second observation points are inserted at set intervals on the second road segment to be matched. The number of second observation points represents the length of the second road segment to be matched. The length of the first road segment to be matched s1 that can be observed from the second road segment to be matched is the total number of intersection points on the first road segment to be matched s1 where the normals of the second observation points intersect. That is, in this embodiment, the ratio of the length of the first road segment to the length of the second road segment observed from the second road segment to the length of the second road segment is the ratio of the total number of intersection points on the first road segment where the normals of the second observation points intersect to the total number of second observation points. For example, as... Figure 2b As shown, the observation length coverage of the road segment s2 to the road segment s1 is the ratio of the total number of intersection points (17) of the road segment s1 to the normal of the road segment s2 to the total number of second observation points (19) of the road segment s2, which is 89.474%.

[0105] As an alternative implementation method, the Hungarian algorithm can be used to match each first road segment to be matched with each second road segment to be matched, so as to obtain the first and second road segments to be matched that achieve the optimal matching conditions.

[0106] The Hungarian algorithm is an algorithm that seeks the maximum match. In this embodiment, the input parameters of the algorithm are the distance and angle between the two road segments to be matched, as well as the observation coverage between them. During the operation of the algorithm, each first road segment to be matched and the second road segment to be matched will be randomly matched. The final output is the road segment that makes the preset matching conditions optimal (e.g., the distance between them is the smallest, the angle values ​​are the closest, and the observation length coverage between them is the largest).

[0107] As another optional implementation, for any first road segment to be matched, the matching score of each second road segment to be matched with the first road segment is determined according to the preset matching conditions, and the second road segment with the largest score is selected as the second road segment to be matched successfully with the first road segment. This process continues until all first road segments to be matched are traversed, and the second road segments to be matched successfully with each first road segment are obtained.

[0108] In this embodiment, thresholds can be set for each preset matching condition. Each threshold has a corresponding score. For example, an observation length coverage rate of 0-30% corresponds to a score of 1; an observation length coverage rate of 31%-60% corresponds to a score of 2; an observation length coverage rate of 61%-90% corresponds to a score of 3; and an observation length coverage rate of 91%-100% corresponds to a score of 4. Other preset matching conditions are similar. For any first road segment to be matched, the scores corresponding to each preset matching condition of each second road segment to be matched with the first road segment are determined. The scores corresponding to each preset matching condition are then summed to obtain the matching score of the second road segment to be matched with the first road segment. The second road segment with the largest score is selected from the matching scores and is considered a successful match with the first road segment. This process continues until all first road segments to be matched are traversed, resulting in a second road segment that has been successfully matched with each of the first road segments.

[0109] Furthermore, after the matching operation is completed, successfully matched skeleton road segments and their corresponding high-precision road segments are obtained. Since high-precision maps consider a variety of other rich road information during mapping, their accuracy is higher than that of the skeleton lines. Therefore, by replacing the accuracy parameters (such as longitude, latitude, and elevation information) of the successfully matched skeleton road segments with the accuracy parameters of the successfully matched high-precision road segments, the accuracy of the skeleton lines can be effectively improved. When using these accuracy-corrected skeleton lines for subsequent matching or mapping, more accurate matching and mapping results can be obtained.

[0110] Building upon the previous embodiments, this embodiment, assuming no connectivity issues with the high-precision map, corrects the accuracy of the skeleton lines at corresponding locations using the high-precision map's accuracy, resulting in more precise skeleton lines. Particularly in tunnel scenarios where skeleton lines reside within the same cell, the method provided in this embodiment can accurately correct the accuracy of tunnel skeleton lines.

[0111] Example 3

[0112] Figure 3 This is a structural block diagram of a high-precision map connectivity verification device provided in Embodiment 3 of the present invention, as shown below. Figure 3 As shown, the device includes: a matching module 310 and a connectivity determination module 320, wherein,

[0113] The matching module 310 is configured to extract the road centerline of the high-precision map to be verified for connectivity, and match the road centerline with the skeleton line at the corresponding location. The matching content includes road label information, road location information and road angle information. The skeleton line is a traditional map obtained by producing open source maps for creating the high-precision map.

[0114] The connectivity determination module 320 is configured to determine that the connectivity of the high-precision map to be verified is not missing if all matching contents of the road centerline and the corresponding skeleton line are successfully matched.

[0115] In the matching results, if there are road segments in the skeleton line that are not matched with the road centerline, it is determined whether the unmatched road segments are road connecting segments, wherein both ends of the road connecting segment have other road segments that are topologically connected to the road connecting segment.

[0116] If the unmatched road segment is not a road connection segment, then it is determined that the connectivity of the high-precision map to be verified is not missing.

[0117] If the unmatched road segment is a road connection segment, it is determined that the connectivity of the high-precision map to be verified is missing.

[0118] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0119] The re-verification module is configured to reconstruct the high-precision map with missing connectivity, and use the reconstructed high-precision map as the new high-precision map to be verified for connectivity. It then returns to perform the operation of extracting the road centerlines from the new high-precision map to be verified for connectivity, and matching the road centerlines with the corresponding skeleton lines, until it is determined that the connectivity of the high-precision map to be verified is not missing.

[0120] Optionally, the matching module is specifically configured as follows:

[0121] Extract the road centerlines from the high-precision map whose connectivity needs to be verified;

[0122] Determine the corresponding road range for each segment in the skeleton line based on the road width of each segment.

[0123] Match the matching contents corresponding to each high-precision road segment in the road centerline with the matching contents corresponding to each road range in the skeleton line.

[0124] Optionally, the high-precision map is constructed at the cell level, with overlapping areas between adjacent cells. Each cell is divided according to different high-precision road segments in the high-precision map.

[0125] Correspondingly, the connectivity of high-precision maps is verified at the cell level;

[0126] Accordingly, after connectivity verification has been completed for all high-precision maps in each cell, the apparatus provided in this embodiment of the invention further includes:

[0127] The merging module is configured to merge cells according to overlapping road segments to obtain a complete high-precision map.

[0128] Optionally, if there are no missing connections in the high-precision map of each cell, the apparatus provided in this embodiment of the invention further includes:

[0129] The matching object segmentation module is configured to, for the first road segment in the successfully matched road centerline and the second road segment in the corresponding matched skeleton line, merge the straight road segments with topological connections in the successfully matched road centerline segment into the same first road segment to be matched, and merge the straight road segments with topological connections in the corresponding matched skeleton line segment into the same second road segment to be matched.

[0130] The road matching module is configured to match each first road segment to be matched with each second road segment to be matched, to obtain the first and second road segments to be matched successfully with the preset matching conditions. The preset matching conditions include distance information, angle information, and observation length coverage. The observation length coverage includes the ratio of the length of the second road segment observed by the first road segment to the length of the first road segment, and the ratio of the length of the first road segment observed by the second road segment to the length of the second road segment.

[0131] The skeleton line accuracy correction module is configured to replace the accuracy parameter value of the second matching road segment that has been successfully matched with the accuracy parameter value of the first matching road segment that has been successfully matched with it. The accuracy parameters include: longitude information, latitude information and elevation information.

[0132] Optionally, the road matching module may include:

[0133] The first matching unit is configured to match each first road segment to be matched with each second road segment to be matched based on the Hungarian algorithm, so as to obtain the first and second road segments to be matched that achieve the optimal matching conditions.

[0134] or,

[0135] The second matching unit is configured to, for any first road segment to be matched, determine the matching score of each second road segment to be matched with the first road segment according to the preset matching conditions, and select the second road segment corresponding to the largest score from the matching scores as the second road segment to be matched successfully matched with the first road segment, until all first road segments to be matched have been traversed, and the second road segment to be matched successfully matched with each first road segment is obtained.

[0136] The high-precision map connectivity verification device provided in this embodiment of the invention can execute the high-precision map connectivity verification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in the above embodiments can be found in the high-precision map connectivity verification method provided in any embodiment of the invention.

[0137] Example 4

[0138] Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention, as shown below. Figure 4 As shown, the electronic device includes:

[0139] Memory 510 storing executable program code;

[0140] Processor 520 coupled to memory 510;

[0141] The processor 520 calls the executable program code stored in the memory 510 to execute the connectivity verification method of the high-precision map provided in any embodiment of the present invention.

[0142] Example 6

[0143] Based on the above embodiments, another embodiment of the present invention provides a vehicle that includes the device as described in any of the above embodiments, or includes electronic equipment as described above.

[0144] Figure 5 This is a schematic diagram of a vehicle provided in Embodiment 5 of the present invention. Figure 5As shown, the vehicle includes a speed sensor 61, an ECU (Electronic Control Unit) 62, a GPS (Global Positioning System) positioning device 63, and a T-Box (Telematics Box) 64. The speed sensor 61 measures the vehicle speed and uses this speed as an empirical speed for model training; the GPS positioning device 63 obtains the vehicle's current geographical location; the T-Box 64 can act as a gateway to communicate with the server; and the ECU 62 can execute the connectivity verification method for the aforementioned high-precision map.

[0145] In addition, the vehicle may also include: a V2X (Vehicle-to-Everything) module 65, a radar 66, and a camera 67. The V2X module 65 is used to communicate with other vehicles, roadside equipment, etc.; the radar 66 or camera 67 is used to perceive road environment information in front and / or other directions to obtain raw point cloud data; the radar 66 and / or camera 67 can be configured at the front and / or rear of the vehicle.

[0146] Based on the above method embodiments, another embodiment of the present invention provides a storage medium storing executable instructions, which, when executed by a processor, cause the processor to implement the connectivity verification method for high-precision maps as described in any of the above embodiments.

[0147] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0148] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions 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 connectivity verification method of a high-precision map, characterized by, The method comprises the following steps: extracting a road center line of a high-precision map to be verified for connectivity, and matching the road center line with a skeleton line at a corresponding position, wherein the matching content comprises road label information, road position information and road angle information, and the skeleton line is a traditional map used to create the high-precision map obtained through open-source map production; if each item of the matching content of the road center line and the skeleton line at the corresponding position is successfully matched, it is determined that the connectivity of the high-precision map to be verified is not missing; in the matching result, if there is a road segment in the skeleton line that is not matched with the road center line, it is judged whether the road segment that is not matched is a road connection road segment, wherein both ends of the road connection road segment exist other road segments topologically connected to the road connection road segment; if the road segment that is not matched is not a road connection road segment, it is determined that the connectivity of the high-precision map to be verified is not missing; if the road segment that is not matched is a road connection road segment, it is determined that the connectivity of the high-precision map to be verified is missing; The method further comprises: re-mapping the high-precision map with missing connectivity, and taking the high-precision map obtained by re-mapping as a new high-precision map to be verified for connectivity, and returning to perform the operation of extracting the road center line of the new high-precision map to be verified for connectivity, and matching the road center line with the skeleton line at the corresponding position, until it is determined that the connectivity of the high-precision map to be verified is not missing.

2. The method of claim 1, wherein, The matching of the road center line with the skeleton line at the corresponding position comprises: determining the road range corresponding to each road segment in the skeleton line according to the road width of each road segment in the skeleton line; matching each item of matching content corresponding to each high-precision road segment in the road center line with each item of matching content corresponding to the road range of each road segment in the skeleton line.

3. The method of claim 1, wherein, The high-precision map is mapped according to the granularity of cells, and there is an overlapping area between adjacent cells, and each cell is divided according to different high-precision road segments in the high-precision map; Correspondingly, the connectivity of the high-precision map is verified according to the granularity of cells; Correspondingly, after the high-precision map of each cell is verified for connectivity, the method further comprises: merging each cell according to the overlapping road segments to obtain a complete high-precision map.

4. The method of claim 3, wherein, In the case that the connectivity of the high-precision map of each cell is not missing, the method further comprises: for a first road segment in the road center line that is successfully matched, and a second road segment in the skeleton line that is successfully matched, merging straight road segments that exist topological connection in the first road segment that is successfully matched into a same first road segment to be matched, and merging straight road segments that exist topological connection in the second road segment that is successfully matched into a same second road segment to be matched; match each first to-be-matched road segment with each second to-be-matched road segment to obtain a first to-be-matched road segment and a second to-be-matched road segment that are successfully matched so that a preset matching condition is optimal, wherein the preset matching condition includes distance information, angle information, and observation length coverage, and the observation length coverage includes a ratio value of a length of the second to-be-matched road segment observed by the first to-be-matched road segment to a length of the first to-be-matched road segment and a ratio value of a length of the first to-be-matched road segment observed by the second to-be-matched road segment to a length of the second to-be-matched road segment; replace an accuracy parameter value of the second to-be-matched road segment that is successfully matched with an accuracy parameter value of the first to-be-matched road segment that is successfully matched with the second to-be-matched road segment, wherein the accuracy parameter includes longitude information, latitude information, and elevation information.

5. The method of claim 4, wherein, The matching, to obtain a first to-be-matched road segment and a second to-be-matched road segment that are successfully matched so that a preset matching condition is optimal, includes: The first implementation manner includes: matching, based on a Hungarian algorithm, each first to-be-matched road segment with each second to-be-matched road segment to obtain a first to-be-matched road segment and a second to-be-matched road segment that are successfully matched so that a preset matching condition is optimal; The second implementation manner includes: for each first to-be-matched road segment, determining a matching score value of each second to-be-matched road segment that is matched with the first to-be-matched road segment according to the preset matching condition, and selecting, from the matching score values, a second to-be-matched road segment corresponding to a maximum score value as a second to-be-matched road segment that is successfully matched with the first to-be-matched road segment, until all first to-be-matched road segments are traversed to obtain second to-be-matched road segments that are successfully matched with the first to-be-matched road segments, respectively.

6. A high-precision map connectivity verification device, characterized by, The matching module is configured to extract a road center line of a high-precision map to be verified for connectivity and match the road center line with a skeleton line at a corresponding position, wherein matching content includes road label information, road position information, and road angle information, and the skeleton line is a traditional map used to create the high-precision map and is produced by an open-source map; The connectivity determination module is configured to determine that connectivity of the high-precision map to be verified does not have a defect if each item of matching content of the road center line and the skeleton line at the corresponding position is successfully matched. In the matching result, if there is a road segment in the skeleton line that is not matched with the road center line, it is determined whether the road segment that is not matched is a road connection road segment, wherein both ends of the road connection road segment have other road segments that are topologically connected to the road connection road segment. If the road segment that is not matched is not a road connection road segment, it is determined that the connectivity of the high-precision map to be verified does not have a defect. If the road segment that is not matched is a road connection road segment, it is determined that the connectivity of the high-precision map to be verified has a defect. The apparatus further includes ​ The rechecking module is configured to re-map the high-precision map with missing connectivity, and take the high-precision map obtained by re-mapping as a new high-precision map to be checked for connectivity, and return to perform the operation of extracting the road center line of the new high-precision map to be checked for connectivity, and matching the road center line with the skeleton line at the corresponding position until it is determined that the high-precision map to be checked for connectivity has no missing connectivity.

7. The apparatus of claim 6, wherein, The matching module is specifically configured to: extract the road center line of the high-precision map to be checked for connectivity; determine the road range corresponding to each road segment in the skeleton line according to the road width of each road segment in the skeleton line; match each item of matching content corresponding to each high-precision road segment in the road center line with each item of matching content corresponding to the road range of each road segment in the skeleton line.

8. The apparatus of claim 6, wherein, The high-precision map is mapped according to the granularity of cells, and there is an overlapping area between adjacent cells, and each cell is divided according to different high-precision road segments in the high-precision map; Correspondingly, the connectivity of the high-precision map is checked according to the granularity of the cells; Correspondingly, after the high-precision map of each cell is checked for connectivity, the device further comprises: The merging module is configured to merge each cell according to the overlapping road segments to obtain a complete high-precision map.

9. The apparatus of claim 8, wherein, In the case that the connectivity of the high-precision map of each cell has no missing, the device further comprises: The matching object division module is configured to, for the first road segment in the road center line that is successfully matched and the second road segment in the skeleton line that is successfully matched, merge straight road segments that have topological connections in the first road segment in the road center line that is successfully matched into a same first road segment to be matched, and merge straight road segments that have topological connections in the second road segment in the skeleton line that is successfully matched into a same second road segment to be matched; The road matching module is configured to match each first road segment to be matched with each second road segment to be matched to obtain a first road segment to be matched and a second road segment to be matched that are successfully matched and make a preset matching condition optimal, wherein the preset matching condition includes distance information, angle information, and observation length coverage rate, and the observation length coverage rate includes a proportional value of the length of the second road segment to be matched observed by the first road segment to be matched to the length of the first road segment to be matched, and a proportional value of the length of the first road segment to be matched observed by the second road segment to be matched to the length of the second road segment to be matched; The skeleton line accuracy correction module is configured to replace the accuracy parameter value of the second road segment to be matched that is successfully matched with the accuracy parameter value of the first road segment to be matched that is successfully matched, wherein the accuracy parameter includes longitude information, latitude information, and elevation information.

10. The apparatus of claim 9, wherein, The road matching module includes: The first matching unit is configured to match each first road segment to be matched with each second road segment to be matched based on the Hungarian algorithm to obtain a first road segment to be matched and a second road segment to be matched that are successfully matched and make a preset matching condition optimal; or, A second matching unit is configured to, for each first to-be-matched road segment, determine a matching score value of each second to-be-matched road segment matched with the first to-be-matched road segment according to the preset matching condition, and select a second to-be-matched road segment corresponding to a maximum score value from the matching score values as a second to-be-matched road segment successfully matched with the first to-be-matched road segment, until all the first to-be-matched road segments are traversed, and a second to-be-matched road segment successfully matched with each first to-be-matched road segment is obtained.

11. An electronic device, comprising: comprising: one or more processors; a memory device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-5.

12. A storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the method according to any one of claims 1-5.

13. A vehicle characterized by comprising: The vehicle comprises the device according to any one of claims 6-10, or the electronic device according to claim 11.

Images