High-precision map change discovery data verification method and device and electronic equipment

Abstract

The invention relates to the field of high-precision map information verification, and provides a high-precision map change discovery data verification method and apparatus, and an electronic device, which utilize the characteristics of small data volume, high processing efficiency and direct reflection of road traffic information change of a vehicle track to carry out directional information mining based on vehicle track data before and after the change. Determining first verification basic reference data according to the mining result, determining second verification basic reference data according to the mining result of the changed driving track data and a high-precision map, and obtaining change condition information of the road traffic information according to the first verification basic reference data and the second verification basic reference data. And the change condition information is taken as a verification basis, the change discovery data to be verified is verified, and a verification result is obtained, so that automatic, efficient and high-quality verification of the change discovery data is realized, the data processing efficiency is improved, rapid and high-quality updating of the high-precision map can be realized, and meanwhile, the freshness of the high-precision map is improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of high-precision map information verification, and in particular to a high-precision map change discovery data verification method and device, and electronic equipment. Background Art

[0002] With the continuous upgrading of intelligent driving, application terminals such as autonomous driving vehicles are increasingly in need of a low-cost, high-freshness, high-precision map (HD map) that meets the needs of mass production and large-scale coverage. Therefore, in order to achieve timely updates of high-precision maps, existing technologies are exploring the multi-source update technology route. With the development of intelligent connected vehicles, various application terminals have a strong demand for the freshness of HD maps and will also generate a large amount of HD map change discovery data. Roadside terminals and municipal departments will also generate or have corresponding change discovery data. These change discovery data can be used to update HD maps to improve the freshness of HD maps. However, the above change discovery data usually has uneven quality. If you want to ensure the quality of the updated HD map, you must ensure the quality of the above change discovery data source.

[0003] At present, the existing technology usually uses manual verification as the main method to conduct quality inspection on the above change discovery data. However, this process is time-consuming and labor-intensive, which will lead to untimely updates of high-precision maps and affect their freshness. In order to verify the correctness of the above change discovery data, the existing technology has also developed a change discovery data extraction algorithm, including a semi-automatic and semi-manual comparative analysis technology of change discovery data and real-life pictures based on image recognition element extraction, and an automatic cross-validation technology for extracting elements and attribute information based on heterogeneous or different data sources collected at different times, and the extraction results and change discovery data.

[0004] For the semi-automatic and semi-manual comparative analysis technology of change discovery data and real-life pictures based on image recognition element extraction, the homogeneity of its algorithm leads to homogeneity problems when verifying basic data and source change discovery data. On this basis, adding a method of manual comparison of real-life pictures leads to very low efficiency. Therefore, the high-precision map updated by this technology not only falls far short of the requirements of autonomous driving in terms of quality and freshness, but also has a long update time, which cannot meet the timeliness requirements of high-precision map updates.

[0005] For the extraction of feature and attribute information based on heterogeneous data sources or data sources collected at different times, and the automated cross-validation technology for the extraction results and change discovery data, it not only requires a series of process flows such as data source aggregation, fusion processing, and feature extraction, but also consumes a lot of computing power and time, and cannot meet the timeliness of updating high-precision maps. In addition, in order to ensure the quality of the basic verification data, the quality requirements of the data source and the algorithm are very high, and it is difficult to meet the freshness and quality requirements of high-precision maps.

[0006] In summary, when faced with changes in high-precision maps, it is found that the quality of multi-source data is uneven. The existing technology is inefficient in verifying the quality of these data, resulting in a long update time for high-precision maps. Its quality and freshness are also difficult to meet the application requirements of autonomous driving. Summary of the invention

[0007] The present disclosure aims to solve at least one of the problems existing in the prior art and to provide a method and device for verifying change discovery data in a high-precision map, and an electronic device.

[0008] In one aspect of the present disclosure, a method for verifying high-precision map change discovery data is provided, the method comprising:

[0009] Determine the area to be verified of the high-precision map based on the data source range corresponding to the change discovery data to be verified for the high-precision map;

[0010] Based on the to-be-verified area, respectively obtaining a vehicle trajectory data family before the change and a vehicle trajectory data family after the change corresponding to the to-be-verified change discovery data;

[0011] Performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change;

[0012] Determining first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change;

[0013] Determining second verification basic reference data based on the changed vehicle trajectory mining information and the high-precision map;

[0014] Based on the first verification basic parameter data and the second verification basic reference data, a verification result corresponding to the change discovery data to be verified is determined.

[0015] Optionally, the determining the first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change includes:

[0016] The vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change are compared and analyzed according to the preset target. If the information for the same preset target is inconsistent, the corresponding change content is used as the first verification basic reference data.

[0017] Optionally, the determining of the second verification basic reference data based on the mining information of the vehicle trajectory after the change and the high-precision map includes:

[0018] The vehicle trajectory mining information after the change and the high-precision map are compared and analyzed according to the preset target. If the information for the same preset target is inconsistent, the corresponding change content will be used as the second verification basic reference data.

[0019] Optionally, the determining, based on the first verification basic parameter data and the second verification basic reference data, a verification result corresponding to the change discovery data to be verified includes:

[0020] The first verification basic parameter data and the second verification basic reference data are compared and analyzed according to the preset target. If the information for the same preset target is consistent, the change discovery data to be verified is compared and analyzed with the first verification basic parameter data or the second verification basic parameter data to obtain the verification result corresponding to the change discovery data to be verified.

[0021] Optionally, the preset target includes an arrow type; the performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change includes:

[0022] Obtain the entry and exit trajectories of a vehicle at the same intersection;

[0023] Determine the tangent angle corresponding to the trajectory entering the intersection and the trajectory exiting the intersection;

[0024] Based on the tangent angle, determining lane turning information corresponding to the trajectory entering the intersection;

[0025] Based on the lane turning information, arrow type information of the corresponding lane is determined.

[0026] Optionally, the preset target includes a stop line position; the performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change includes:

[0027] For intersections with traffic lights, determine whether there are multiple vehicle trajectories that stay within a preset distance range before the intersection for no more than a preset time, and each of the vehicle trajectories meets the trajectory homogeneity requirements for different time periods. If so, the position close to the intersection within the preset distance range before the intersection is used as the stop line position.

[0028] Optionally, the preset target includes a lane marking type; the performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change includes:

[0029] For adjacent lanes on the same road, multiple vehicle trajectories over multiple days and time periods are extracted;

[0030] According to the lane changing trends indicated by the multiple vehicle trajectories, lane marking type information corresponding to each lane is determined respectively.

[0031] Optionally, before performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change, the verification method further includes:

[0032] The vehicle trajectory data family before the change and the vehicle trajectory data family after the change are fused respectively to achieve lane-level vehicle trajectory fusion.

[0033] Another aspect of the present disclosure provides a high-precision map change discovery data verification device, the verification device comprising:

[0034] A first determination module, configured to determine the area to be verified of the high-precision map based on a data source range corresponding to the change discovery data to be verified for the high-precision map;

[0035] An acquisition module, used for respectively acquiring a vehicle trajectory data group before the change and a vehicle trajectory data group after the change corresponding to the change discovery data to be verified based on the area to be verified;

[0036] A mining module, used to perform directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target, to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change;

[0037] A second determination module, configured to determine first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change;

[0038] A third determination module is used to determine the second verification basic reference data based on the changed vehicle trajectory mining information and the high-precision map;

[0039] The fourth determination module is used to determine the verification result corresponding to the change discovery data to be verified based on the first verification basic parameter data and the second verification basic reference data.

[0040] Another aspect of the present disclosure provides an electronic device, including:

[0041] at least one processor; and,

[0042] a memory communicatively connected to at least one processor; wherein,

[0043] The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor so that the at least one processor can execute the high-precision map change discovery data verification method described above.

[0044] Compared with the prior art, the present invention utilizes the characteristics of vehicle trajectories, which have small data volume, high processing efficiency, and can directly reflect changes in road traffic information. It conducts directional information mining based on a large amount of vehicle trajectory data after changes and vehicle trajectory data before changes, and determines first verification basic reference data according to the mining results. At the same time, second verification basic reference data is determined according to the mining results of the vehicle trajectory data after changes and the high-precision map. According to the first verification basic reference data and the second verification basic reference data, information on changes in road traffic information is obtained, and the change information is used as a verification basis to verify the change discovery data to be verified, and corresponding verification results are obtained, thereby realizing automatic, efficient, and high-quality verification of change discovery data, improving data processing efficiency, and improving the quality of change discovery data used for updating high-precision maps. On the premise of ensuring the quality of high-precision maps, fast and high-quality updates of high-precision maps can be achieved, while improving the freshness of high-precision maps. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0046] Figure 1 A flowchart of a method for verifying high-precision map change discovery data provided by an embodiment of the present disclosure;

[0047] Figure 2 A flowchart of mining arrow type information provided by another embodiment of the present disclosure;

[0048] Figure 3 A schematic diagram of turning angles of a vehicle trajectory provided by another embodiment of the present disclosure;

[0049] Figure 4 A flowchart for mining lane marking type information provided by another embodiment of the present disclosure;

[0050] Figure 5A flowchart of a method for verifying high-precision map change discovery data provided by another embodiment of the present disclosure;

[0051] Figure 6 A schematic diagram of the structure of a high-precision map change discovery data verification device provided by another embodiment of the present disclosure;

[0052] Figure 7 A schematic diagram of an application scenario of a high-precision map change discovery data verification device provided by another embodiment of the present disclosure;

[0053] Figure 8 A schematic structural diagram of an electronic device provided in another embodiment of the present disclosure. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in each embodiment of the present disclosure, many technical details are proposed in order to enable readers to better understand the present disclosure. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed for protection in the present disclosure can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present disclosure. The various embodiments can be combined and referenced with each other without contradiction.

[0055] One embodiment of the present disclosure relates to a method for verifying change discovery data of a high-precision map. The verification method takes into account that vehicle trajectory can directly reflect changes in road traffic information, performs targeted information mining based on a large amount of vehicle trajectory data before and after the change, and uses the information before and after the change to verify the change discovery data, thereby solving the problem of verifying road feature change discovery data and improving the quality and freshness of high-precision map updates.

[0056] like Figure 1 As shown in the figure, the data verification method for high-precision map change discovery includes:

[0057] Step S110, determining the area to be verified of the high-precision map based on the data source range corresponding to the change discovery data to be verified for the high-precision map.

[0058] Specifically, HD map is a high-precision map used for autonomous driving. As a priori data for autonomous driving, it expresses map elements such as roads, lanes, roadside traffic signs and ground signs, and its map accuracy can be down to the centimeter level. Change discovery data for HD maps refers to the content data of changes in road scenes and road element information expressed based on historical versions of HD maps. Change discovery data to be verified refers to change discovery data that needs to be verified for correctness.

[0059] Step S110 mainly extracts the regional scope based on the change discovery data to be verified. Specifically, the data source range, that is, the data source area, corresponding to the change discovery data to be verified can be first determined, and then the area in the high-precision map corresponding to the data source area is used as the area to be verified, thereby realizing the regional scope extraction based on the change discovery data.

[0060] Step S120 , based on the area to be verified, respectively obtaining a vehicle trajectory data set before the change and a vehicle trajectory data set after the change corresponding to the change discovery data to be verified.

[0061] Specifically, the vehicle trajectory refers to a string of driving trajectory points composed of the driving trajectory points of crowd-source vehicles. Among them, each driving trajectory point includes information such as time and geographic coordinates. The vehicle trajectory data family before the change corresponding to the change discovery data to be verified refers to a data set composed of the vehicle trajectories of each crowd-source vehicle whose corresponding high-precision map range is within the range of the area to be verified and generated before the change discovery data to be verified. The vehicle trajectory data family after the change corresponding to the change discovery data to be verified refers to a data set composed of the vehicle trajectories of each crowd-source vehicle whose corresponding high-precision map range is within the range of the area to be verified and generated after the change discovery data to be verified. Both the vehicle trajectory data family before the change and the vehicle trajectory data family after the change can be directly obtained through each crowd-source vehicle.

[0062] In particular, after step S120 and before step S130, in order to improve the accuracy of the vehicle trajectory and realize lane-level vehicle trajectory fusion, the high-precision map change discovery data verification method may also include: separately fusing the vehicle trajectory data family before the change and the vehicle trajectory data family after the change to realize lane-level vehicle trajectory fusion.

[0063] Among them, lane-level vehicle trajectory fusion is to fuse the vehicle trajectories belonging to the same target together through fusion processing technology, so that the measured trajectory is closer to the real trajectory, and the accuracy of the fused trajectory can reach the lane level. The fusion processing may include but is not limited to denoising, clustering, fitting and other processing of trajectory information. After fusing the vehicle trajectory data family before the change and the vehicle trajectory data family after the change, the fused vehicle trajectory corresponding to the vehicle trajectory data family before the change and the vehicle trajectory data family after the change can be obtained.

[0064] Step S130 , performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target, to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change.

[0065] Specifically, directional information mining can be performed based on preset targets corresponding to road feature information. For example, the preset targets may include but are not limited to arrow types, stop line positions, lane marking types, and the like.

[0066] Exemplarily, when the preset target includes an arrow type, step S130 includes: obtaining an entry trajectory and an exit trajectory of a certain vehicle trajectory for the same intersection; determining the tangent angle corresponding to the entry trajectory and the exit trajectory; based on the tangent angle, determining the lane turning information corresponding to the entry trajectory; based on the lane turning information, determining the arrow type information of the corresponding lane.

[0067] Specifically, combined with Figure 2 , assuming that a vehicle trajectory is a fused vehicle trajectory, the turning angle between its entry trajectory and exit trajectory at the same intersection can be used to determine the lane turning information corresponding to its entry trajectory. Then, by mining the lane turning information corresponding to the entry trajectory, the type of the indicator arrow on the corresponding lane, i.e., the lane arrow type, can be determined. Among them, the turning angle can be obtained by calculating the tangent angle between the entry trajectory and the exit trajectory. Figure 3 As shown, assuming that a vehicle trajectory enters, passes through and exits the intersection along the direction of the arrow, the angle α between the tangent of its trajectory entering the intersection and the tangent of its trajectory exiting the intersection is the turning angle between its trajectory entering the intersection and its trajectory exiting the intersection. By judging the size of the angle α and the preset corresponding relationship between the turning angle and the lane steering, the lane steering information corresponding to the trajectory entering the intersection can be obtained.

[0068] For example, assuming that the turning angle α corresponding to a certain vehicle trajectory passing through an intersection is 90 degrees or close to 90 degrees, it means that the lane turning information corresponding to the trajectory entering the intersection of the vehicle trajectory includes a left turn, and the arrow type information of the corresponding lane, i.e., the lane where the trajectory entering the intersection is located, includes a left turn. If the turning angles corresponding to all vehicle trajectories passing through the intersection through the same entry lane only meet 90 degrees or close to 90 degrees, it means that the arrow type information of the entry lane only includes a left turn.

[0069] Assuming that the turning angle α corresponding to a vehicle trajectory passing through an intersection is 180 degrees or close to 180 degrees, it means that the lane turning information corresponding to the trajectory entering the intersection of the vehicle trajectory includes straight driving, and the arrow type information of the corresponding lane, i.e., the lane where the trajectory entering the intersection is located, includes straight driving. If the turning angles corresponding to all vehicle trajectories passing through the intersection through the same entry lane only meet 180 degrees or close to 180 degrees, it means that the arrow type information of the entry lane only includes straight driving.

[0070] Assuming that the turning angle α corresponding to a certain vehicle trajectory passing through an intersection is 270 degrees or close to 270 degrees, it means that the lane turning information corresponding to the trajectory entering the intersection of the vehicle trajectory includes a right turn, and the arrow type information of the corresponding lane, i.e., the lane where the trajectory entering the intersection is located, includes a right turn. If the turning angles corresponding to all vehicle trajectories passing through the intersection through the same entry lane only meet 270 degrees or close to 270 degrees, it means that the arrow type information of the entry lane only includes a right turn.

[0071] If the turning angles corresponding to all vehicle trajectories passing through the intersection using the same entry lane include both 90 degrees or close to 90 degrees and 180 degrees or close to 180 degrees, it means that the arrow type of the entry lane includes both left turn and straight going.

[0072] If the turning angles corresponding to all vehicle trajectories passing through the intersection using the same entry lane include both 90 degrees or close to 90 degrees and 270 degrees or close to 270 degrees, it means that the arrow type of the entry lane includes both left turns and right turns.

[0073] If the turning angles corresponding to all vehicle trajectories passing through the intersection using the same entry lane include both 180 degrees or close to 180 degrees and 270 degrees or close to 270 degrees, it means that the arrow type of the entry lane includes both straight driving and right turning.

[0074] If the turning angles corresponding to all vehicle trajectories passing through the intersection using the same entry lane include an angle of 90 degrees or close to 90 degrees, an angle of 180 degrees or close to 180 degrees, and an angle of 270 degrees or close to 270 degrees, then it means that the arrow type of the entry lane includes left turn, straight go and right turn at the same time.

[0075] Exemplarily, when the preset target includes the stop line position, step S130 includes: for an intersection with a traffic light, determining whether there are multiple vehicle trajectories that stay within a preset distance range in front of the intersection for no more than a preset time, and each vehicle trajectory meets the trajectory homogeneity requirements for different time periods; if so, then the position close to the intersection within the preset distance range in front of the intersection is used as the stop line position.

[0076] Specifically, the preset duration can be set to the red light waiting time of the corresponding traffic light. If there are multiple vehicle trajectories that stay within the preset distance range before the same intersection, and the respective stay time does not exceed the preset time, and the vehicle trajectories meet the homogeneity requirements of trajectories in different time periods, it means that the stay of these vehicle trajectories within the preset distance range of the intersection is a behavior of waiting for the red light. At this time, there is a stop line near the intersection. Among them, the homogeneity of trajectories in different time periods means that the vehicle trajectories in different time periods are in the same or similar state.

[0077] Exemplarily, when the preset target includes lane marking types, step S130 includes: for adjacent lanes included in the same road, extracting multiple vehicle trajectories within multiple days and time periods; and determining the lane marking type information corresponding to each lane according to the lane change trends indicated by the multiple vehicle trajectories.

[0078] Specifically, combined with Figure 4 Assuming that the vehicle trajectory data family before the change and the vehicle trajectory data family after the change have their corresponding fused vehicle trajectories, step S130 can mine the trajectory trends of two adjacent lanes on the same road from the fused vehicle trajectory, and mine the lane marking type information corresponding to different trajectory trends through trajectory trend judgment.

[0079] For example, if multiple vehicle trajectories indicate that the vehicle trajectories corresponding to adjacent lanes on the same road have a regular trend of changing lanes with each other, then the lane marking type between adjacent lanes is considered to be a dotted line. If multiple vehicle trajectories indicate that the vehicle trajectories corresponding to adjacent lanes on the same road have no trend of changing lanes with each other, then the lane marking type between adjacent lanes is considered to be a solid line. If multiple vehicle trajectories indicate that the vehicle trajectories corresponding to adjacent lanes on the same road only have a trend of changing lanes to the right but not to the left, then for the right lane of the two adjacent lanes, the lane marking type on its left side is a dotted line, and the lane marking type on its right side is a solid line, that is, the left is dotted and the right is solid. If multiple vehicle trajectories indicate that the vehicle trajectories corresponding to adjacent lanes on the same road only have a trend of changing lanes to the left but not to the right, then for the left lane of the two adjacent lanes, the lane marking type on its left side is a solid line, and the lane marking type on its right side is a dotted line, that is, the left is solid and the right is dotted.

[0080] Step S140, determining first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change.

[0081] Specifically, step S140 may compare and analyze the vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change, and determine the first verification basic reference data according to the analysis result.

[0082] Exemplarily, step S140 includes: comparing and analyzing the vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change according to a preset target, and if the information for the same preset target is inconsistent, the corresponding change content is used as the first verification basic reference data.

[0083] Specifically, if the information of the vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change is consistent for the same preset target, it means that the corresponding information has not changed. At this time, the corresponding change content cannot be used as the first basic reference data for verification. If the information of the vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change is inconsistent for the same preset target, it means that the corresponding information has changed. At this time, by using the corresponding change content as the first basic reference data for verification, a high-quality reference standard can be provided for the verification of the change discovery data to be verified, which helps to improve the verification quality of the change discovery data to be verified.

[0084] Step S150, determining the second verification basic reference data based on the vehicle trajectory mining information after the change and the high-precision map.

[0085] Specifically, step S150 can compare and analyze the vehicle trajectory mining information after the change and the high-precision map, and determine the second verification basic reference data according to the analysis results.

[0086] Exemplarily, step S150 includes: comparing and analyzing the vehicle trajectory mining information after the change and the high-precision map according to a preset target. If the information for the same preset target is inconsistent, the corresponding change content is used as the second verification basic reference data.

[0087] Specifically, if the vehicle trajectory mining information after the change is consistent with the information of the high-precision map for the same preset target, it means that the corresponding information has not changed. At this time, the corresponding change content cannot be used as the second basic reference data for verification. If the vehicle trajectory mining information after the change is inconsistent with the information of the high-precision map for the same preset target, it means that the corresponding information has changed. At this time, by using the corresponding change content as the second basic reference data for verification, it can further provide a high-quality reference standard for the verification of the change discovery data to be verified, which helps to improve the verification quality of the change discovery data to be verified.

[0088] Step S160, based on the first verification basic parameter data and the second verification basic reference data, determine the verification result corresponding to the change discovery data to be verified.

[0089] Specifically, step S160 can compare and analyze the first verification basic parameter data and the second verification basic reference data according to different preset targets, determine the final verification basic reference data based on the analysis results, and then use the final verification basic reference data to verify the change discovery data to obtain the corresponding verification result.

[0090] Exemplarily, step S160 includes: comparing and analyzing the first verification basic parameter data and the second verification basic reference data according to a preset target; if the information for the same preset target is consistent, comparing and analyzing the change discovery data to be verified with the first verification basic parameter data or the second verification basic parameter data to obtain a verification result corresponding to the change discovery data to be verified.

[0091] Specifically, if the first verification basic parameter data and the second verification basic reference data are consistent with each other in terms of information for the same preset target, it means that the first verification basic parameter data or the second verification basic reference data are credible verification basic data. At this time, by comparing and analyzing the change discovery data to be verified with the first verification basic parameter data or the second verification basic reference data according to the preset target, the authenticity of the relevant change data in the change discovery data to be verified can be determined, and finally the verification result of the change discovery data to be verified can be obtained.

[0092] If the first verification basic parameter data and the second verification basic reference data are inconsistent with each other for the same preset target, it means that the first verification basic parameter data and the second verification basic reference data are unreliable. At this time, it is necessary to re-mine the information for the area to be verified in the high-precision map to regain the first verification basic reference data and the second verification basic reference data.

[0093] Different from the existing methods based on extracting and cross-validating elements of original perception data sources and the manual verification methods based on real-life pictures, the high-precision map change discovery data verification method provided by the embodiments of the present disclosure utilizes the characteristics of vehicle trajectory data that is small in data volume, high in processing efficiency, and can directly reflect changes in road traffic information. It performs directional information mining based on a large amount of vehicle trajectory data after changes and vehicle trajectory data before changes, determines first verification basic reference data based on the mining results, and determines second verification basic reference data based on the mining results of the vehicle trajectory data after changes and the high-precision map. Information on changes in road traffic information is obtained based on the first verification basic reference data and the second verification basic reference data. The change information is used as a verification basis to verify the change discovery data to obtain corresponding verification results, thereby realizing automated, efficient, and high-quality verification of change discovery data, improving data processing efficiency, and improving the quality of change discovery data used for updating high-precision maps. It can achieve fast and high-quality updates of high-precision maps while ensuring the quality of high-precision maps, while improving the freshness of high-precision maps.

[0094] In order to enable those skilled in the art to better understand the above implementation, a specific example is provided below for illustration.

[0095] like Figure 5 As shown, a method for verifying high-precision map change discovery data includes the following steps:

[0096] 1. Perform regional range extraction based on the data source range of the change discovery data for the high-precision map, determine the area to be verified in the high-precision map, and obtain the vehicle trajectory data family based on the area to be verified, including the vehicle trajectory data family before the change and the vehicle trajectory data family after the change corresponding to the change discovery data.

[0097] 2. Lane-level trajectory fusion: The vehicle trajectory data family before the change and the vehicle trajectory data family after the change are fused separately to obtain the lane-level vehicle trajectories corresponding to the vehicle trajectory data family before the change and the vehicle trajectory data family after the change, respectively, to achieve the fusion of the vehicle trajectory before the change and the vehicle trajectory after the change.

[0098] 3. Directional mining of trajectory information: Directional mining is performed on the fused vehicle trajectory before the change and the fused vehicle trajectory after the change according to the road surface element information to obtain the corresponding mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change, wherein the road surface element information can be divided into arrow type, stop line position, and lane marking type. The mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change respectively include the arrow type information, stop line position information, and lane marking type information corresponding to the fused vehicle trajectory before the change and the fused vehicle trajectory after the change.

[0099] 4. Compare and analyze the vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change, and compare the arrow type information, stop line position information, and lane marking type information respectively: if the information before and after the change is consistent, it is considered that the relevant information has not changed; if the information before and after the change is inconsistent, it is considered that the relevant information has changed, and the change content is obtained as the basic reference data for verification 1.

[0100] 5. Synchronously, compare and analyze the mining information of the vehicle trajectory after the change with the corresponding information in the high-precision map: if the two information are consistent, it is considered that the corresponding information has not changed; if the two information are inconsistent, it is considered that the corresponding information has changed, and the change content is obtained as the basic reference data for verification 2. Among them, the change content can be the change of arrow type, the change of stop line position, and the change of lane marking type.

[0101] 6. Compare and analyze the verification basic reference data 1 and the verification basic reference data 2: If the corresponding information of the two is inconsistent, the relevant area will be treated as a pending area for re-information mining; if the corresponding information of the two is consistent, the verification basic reference data 1 and the verification basic reference data 2 are considered to be credible verification basic data, and the change discovery data to be verified will be compared and analyzed with the verification basic reference data 1 or the verification basic reference data 2 to obtain the verification result of the change discovery data.

[0102] Another embodiment of the present disclosure relates to a high-precision map change discovery data verification device, such as Figure 6 As shown, it includes a first determination module 610, an acquisition module 620, a mining module 630, a second determination module 640, a third determination module 650, and a fourth determination module 660.

[0103] The first determination module 610 is used to determine the area to be verified of the high-precision map based on the data source range corresponding to the change discovery data to be verified for the high-precision map.

[0104] The acquisition module 620 is used to respectively acquire a vehicle trajectory data set before the change and a vehicle trajectory data set after the change corresponding to the change discovery data to be verified based on the area to be verified.

[0105] The mining module 630 is used to perform directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on preset targets to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change.

[0106] The second determination module 640 is used to determine the first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change.

[0107] The third determination module 650 is used to determine the second verification basic reference data based on the vehicle trajectory mining information and the high-precision map after the change.

[0108] The fourth determination module 660 is used to determine the verification result corresponding to the change discovery data to be verified based on the first verification basic parameter data and the second verification basic reference data.

[0109] In particular, the high-precision map change discovery data verification device may further include a fusion module. The fusion module is used to fuse the vehicle trajectory data family before the change and the vehicle trajectory data family after the change, respectively, to achieve lane-level vehicle trajectory fusion.

[0110] The specific implementation method of the high-precision map change discovery data verification device provided in the embodiment of the present disclosure can be found in the high-precision map change discovery data verification method provided in the embodiment of the present disclosure, and will not be repeated here.

[0111] Figure 7 The application scenario of the high-precision map change discovery data verification device is shown. Figure 7 As shown, the change discovery data including arrow type information, stop line position information, lane marking type information, vehicle trajectory before and after the change, and high-precision map are input into a high-precision map change discovery data verification device, i.e., a source change discovery data authenticity verification device, to obtain a verification result of the change discovery data, which includes the authenticity of the arrow type change, the stop line position change, and the lane marking type change in the change discovery data.

[0112] Compared with the prior art, the high-precision map change discovery data verification device provided by the disclosed embodiment utilizes the characteristics of vehicle trajectory data of small data volume, high processing efficiency, and the ability to directly reflect changes in road traffic information. It performs targeted information mining based on a large amount of vehicle trajectory data after changes and vehicle trajectory data before changes, and determines first verification basic reference data according to the mining results. At the same time, second verification basic reference data is determined according to the mining results of the vehicle trajectory data after changes and the high-precision map. According to the first verification basic reference data and the second verification basic reference data, change information of road traffic information is obtained, and the change information is used as the verification basis to verify the change discovery data to obtain corresponding verification results. This realizes automatic, efficient, and high-quality verification of change discovery data, improves data processing efficiency, and improves the quality of change discovery data used for updating high-precision maps. On the premise of ensuring the quality of high-precision maps, it can realize fast and high-quality updates of high-precision maps and improve the freshness of high-precision maps.

[0113] Another embodiment of the present disclosure relates to an electronic device, such as Figure 8 As shown, including:

[0114] at least one processor 801; and,

[0115] A memory 802 is communicatively connected to at least one processor 801; wherein,

[0116] The memory 802 stores instructions that can be executed by at least one processor 801, and the instructions are executed by at least one processor 801 so that the at least one processor 801 can execute the high-precision map change discovery data verification method described in the above embodiment.

[0117] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0118] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0119] Those skilled in the art will appreciate that the above-mentioned embodiments are specific embodiments for implementing the present disclosure, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present disclosure.

Claims

1. A method for verifying high-precision map change discovery data, characterized in that: The verification method includes: Determine the area to be verified of the high-precision map based on the data source range corresponding to the change discovery data to be verified for the high-precision map; Based on the to-be-verified area, respectively obtaining a vehicle trajectory data family before the change and a vehicle trajectory data family after the change corresponding to the to-be-verified change discovery data; Performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change; Determining first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change; Determining second verification basic reference data based on the changed vehicle trajectory mining information and the high-precision map; Based on the first verification basic parameter data and the second verification basic reference data, a verification result corresponding to the change discovery data to be verified is determined.

2. The verification method according to claim 1, characterized in that: The determining of the first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change includes: The vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change are compared and analyzed according to the preset target. If the information for the same preset target is inconsistent, the corresponding change content is used as the first verification basic reference data.

3. The verification method according to claim 1, characterized in that: The determining of the second verification basic reference data based on the changed vehicle trajectory mining information and the high-precision map includes: The vehicle trajectory mining information after the change and the high-precision map are compared and analyzed according to the preset target. If the information for the same preset target is inconsistent, the corresponding change content will be used as the second verification basic reference data.

4. The verification method according to claim 1, characterized in that: The determining, based on the first verification basic parameter data and the second verification basic reference data, a verification result corresponding to the change discovery data to be verified includes: The first verification basic parameter data and the second verification basic reference data are compared and analyzed according to the preset target. If the information for the same preset target is consistent, the change discovery data to be verified is compared and analyzed with the first verification basic parameter data or the second verification basic parameter data to obtain the verification result corresponding to the change discovery data to be verified.

5. The verification method according to claim 1, characterized in that: The preset target includes an arrow type; the directional information mining is performed on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain the corresponding vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change, including: Obtain the entry and exit trajectories of a vehicle at the same intersection; Determine the tangent angle corresponding to the trajectory entering the intersection and the trajectory exiting the intersection; Based on the tangent angle, determining lane turning information corresponding to the trajectory entering the intersection; Based on the lane turning information, arrow type information of the corresponding lane is determined.

6. The verification method according to claim 1, characterized in that: The preset target includes the stop line position; the directional information mining is performed on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain the corresponding vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change, including: For intersections with traffic lights, determine whether there are multiple vehicle trajectories that stay within a preset distance range before the intersection for no more than a preset time, and each of the vehicle trajectories meets the trajectory homogeneity requirements for different time periods. If so, the position close to the intersection within the preset distance range before the intersection is used as the stop line position.

7. The verification method according to claim 1, characterized in that: The preset target includes a lane marking type; the directional information mining is performed on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on the preset target to obtain the corresponding vehicle trajectory mining information before the change and the vehicle trajectory mining information after the change, including: For adjacent lanes on the same road, multiple vehicle trajectories over multiple days and time periods are extracted; According to the lane changing trends indicated by the multiple vehicle trajectories, lane marking type information corresponding to each lane is determined respectively.

8. The verification method according to any one of claims 1 to 7, characterized in that: Before performing directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change, the verification method further includes: The vehicle trajectory data family before the change and the vehicle trajectory data family after the change are fused respectively to achieve lane-level vehicle trajectory fusion.

9. A high-precision map change discovery data verification device, characterized in that: The verification device comprises: A first determination module, configured to determine the area to be verified of the high-precision map based on a data source range corresponding to the change discovery data to be verified for the high-precision map; An acquisition module, used for respectively acquiring a vehicle trajectory data group before the change and a vehicle trajectory data group after the change corresponding to the change discovery data to be verified based on the area to be verified; A mining module, used to perform directional information mining on the vehicle trajectory data family before the change and the vehicle trajectory data family after the change based on a preset target, to obtain corresponding vehicle trajectory mining information before the change and vehicle trajectory mining information after the change; A second determination module, configured to determine first verification basic reference data based on the mining information of the vehicle trajectory before the change and the mining information of the vehicle trajectory after the change; A third determination module is used to determine the second verification basic reference data based on the changed vehicle trajectory mining information and the high-precision map; The fourth determination module is used to determine the verification result corresponding to the change discovery data to be verified based on the first verification basic parameter data and the second verification basic reference data.

10. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the high-precision map change discovery data verification method described in any one of claims 1 to 8.

Images