Multi-trajectory mapping method and apparatus, and vehicle and storage medium

Abstract

A multi-trajectory mapping method and apparatus, and a vehicle and a storage medium. The method comprises: acquiring key frame data corresponding to a plurality of historical travelling trajectories of a target vehicle on a preset travelling route, wherein the key frame data comprises vehicle pose information in the historical travelling trajectories and at least one map feature corresponding to the vehicle pose information (S11); performing alignment processing on the key frame data (A1, A2, A3, A4, B1, B2, B3, B4, B5, B6) corresponding to different historical traveling trajectories (A, B), so as to obtain a plurality of key frame pairing results (A1-B2, A2-B3, A3-B4, A4-B6) (S12); performing feature association processing on the plurality of key frame pairing results (A1-B2, A2-B3, A3-B4, A4-B6), so as to obtain a feature association result, wherein the feature association result is used for determining a feature correspondence among the plurality of key frame pairing results (S13); and performing fusion processing on the basis of the feature association result, so as to generate target map data (S14). The method solves the technical problems in the related art of low mapping efficiency and poor accuracy during offline mapping.

Description

Multi-trajectory mapping methods, devices, vehicles, and storage media

[0001] This application claims priority to Chinese Patent Application No. 2024112168967, filed with the State Intellectual Property Office of China on August 30, 2024, entitled "Multi-trajectory mapping method, apparatus, vehicle and storage medium", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of autonomous driving technology, specifically to a multi-trajectory mapping method, device, vehicle, and storage medium. Background Technology

[0003] Offline mapping can improve positioning accuracy, accelerate path planning, enhance safety, and reduce costs, providing strong support for the development and application of intelligent driving technology. Traditional offline mapping methods using a single-shot route memory suffer from problems such as missed detections and false detections of mapping elements. Repeatedly collecting information from the same route for mapping can alleviate these issues. However, when using information collected multiple times for offline mapping, the differences in time, road conditions, and observed areas each time, along with issues such as unstable perception and positioning signals, make it difficult to accurately calculate the correlations between multiple data collections, further impacting the efficiency and accuracy of offline mapping.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a multi-track mapping method, apparatus, vehicle, and storage medium to at least solve the technical problems of low mapping efficiency and poor accuracy in offline mapping in related technologies.

[0006] According to one embodiment of this application, a multi-trajectory mapping method is provided, comprising: acquiring keyframe data corresponding to multiple historical driving trajectories of a target vehicle on a preset driving route, wherein the keyframe data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information; performing alignment processing on the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results; performing feature association processing on the multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between the multiple keyframe pairing results; and performing fusion processing based on the feature association results to generate target map data.

[0007] In one embodiment, the multiple historical driving trajectories include at least a first driving trajectory and a second driving trajectory, and the keyframe data includes at least a plurality of first keyframes corresponding to the first driving trajectory and a plurality of second keyframes corresponding to the second driving trajectory. Alignment processing is performed on the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results, including: performing position association processing on the plurality of first keyframes and the plurality of second keyframes to obtain position association results, wherein the position association results are used to represent the initial pairing results of the first keyframes and the second keyframes; performing spatiotemporal association processing based on the position association results to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent the keyframe chain that satisfies the preset spatiotemporal continuity condition; and determining multiple keyframe pairing results based on the spatiotemporal association results.

[0008] In one embodiment, feature association processing is performed on multiple keyframe pairing results to obtain feature association results, including: obtaining vehicle pose information and lane line information based on multiple keyframe pairing results; performing a first association processing using the vehicle pose information and lane line information to obtain lane line association results, wherein the lane line association results are used to represent the lane line correspondence in the keyframe pairing results; and performing a second association processing using the lane line association results to obtain feature association results.

[0009] In one embodiment, the first association processing using vehicle pose information and lane line information to obtain lane line association results includes: performing feature matching using lane line information to obtain a first matching result; performing detection processing on the first matching result based on vehicle pose information to obtain a target detection result, wherein the target detection result is used to determine whether the lane line association distance between adjacent keyframes meets a preset distance condition; and determining the lane line association result based on the target detection result.

[0010] In one embodiment, the second association processing using the lane line association results to obtain the feature association results includes: adjusting the vehicle pose information based on the lane line association results to obtain the pose adjustment results; and performing the second association processing based on at least one map feature in the pose adjustment results and keyframe pairing results to obtain the feature association results.

[0011] In one embodiment, the second association processing based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain the feature association result includes: performing feature matching based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a second matching result; performing pose update processing based on the second matching result to obtain a pose update result; and determining the feature association result based on the pose update result in response to the pose update result satisfying a preset convergence condition.

[0012] In one embodiment, generating target map data by performing fusion processing based on feature association results includes: filtering the feature association results using preset filtering conditions to obtain filtering results, wherein the preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results; and performing fusion processing based on the filtering results to generate target map data.

[0013] In one embodiment, the keyframe data is extracted from the historical driving trajectory using a trajectory analysis algorithm.

[0014] In one embodiment, the step of aligning the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results includes: aligning the keyframe data corresponding to different historical driving trajectories according to the similarity between them to obtain multiple keyframe pairing results.

[0015] In one embodiment, the step of aligning the keyframe data corresponding to different historical driving trajectories based on the similarity between the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results includes: calculating the similarity between each pair of keyframes in the keyframe data corresponding to different historical driving trajectories; sorting the keyframes according to the similarity results; and selecting the keyframes with high similarity for pairing to obtain multiple keyframe pairing results.

[0016] In one embodiment, the feature association processing includes a coarse association process and a fine association process, wherein the coarse association process uses only lane line information in the map features for association, and the fine association process uses all map features for association.

[0017] According to one embodiment of this application, a multi-trajectory mapping device is also provided, comprising: an acquisition module, configured to acquire keyframe data corresponding to multiple historical driving trajectories of a target vehicle on a preset driving route, wherein the keyframe data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information; a processing module, configured to perform alignment processing on the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results; an association module, configured to perform feature association processing on the multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between the multiple keyframe pairing results; and a generation module, configured to perform fusion processing based on the feature association results to generate target map data.

[0018] In one embodiment, the processing module is further configured to: perform position association processing on multiple first keyframes and multiple second keyframes to obtain position association results, wherein the position association results are used to represent the initial pairing results of the first keyframes and the second keyframes; perform spatiotemporal association processing based on the position association results to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent the keyframe chain that satisfies the preset spatiotemporal continuity condition; and determine the pairing results of multiple keyframes based on the spatiotemporal association results.

[0019] In one embodiment, the association module is further configured to: obtain vehicle pose information and lane line information based on multiple keyframe pairing results; perform a first association process using the vehicle pose information and lane line information to obtain a lane line association result, wherein the lane line association result is used to represent the lane line correspondence in the keyframe pairing results; and perform a second association process using the lane line association result to obtain a feature association result.

[0020] In one embodiment, the association module is further configured to: perform feature matching using lane line information to obtain a first matching result; perform detection processing on the first matching result based on vehicle pose information to obtain a target detection result, wherein the target detection result is used to determine whether the lane line association distance between adjacent keyframes meets a preset distance condition; and determine the lane line association result based on the target detection result.

[0021] In one embodiment, the association module is further configured to: adjust the vehicle pose information based on the lane line association result to obtain a pose adjustment result; and perform a second association process based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a feature association result.

[0022] In one embodiment, the association module is further configured to: perform feature matching based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a second matching result; perform pose update processing based on the second matching result to obtain a pose update result; and determine the feature association result based on the pose update result in response to the pose update result satisfying a preset convergence condition.

[0023] In one embodiment, the generation module is further configured to: filter the feature association results using preset filtering conditions to obtain filtering results, wherein the preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results; and perform fusion processing based on the filtering results to generate target map data.

[0024] According to one embodiment of this application, a vehicle is provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes any one of the multi-track mapping methods in various embodiments of this application during runtime.

[0025] According to one embodiment of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the storage medium is located to execute any one of the multi-trajectory mapping methods in various embodiments of this application.

[0026] According to one embodiment of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the multi-trajectory mapping method of any one of the various embodiments of this application.

[0027] According to one embodiment of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the multi-trajectory mapping method of any one of the various embodiments of this application.

[0028] According to one embodiment of this application, a computer program is also provided, which, when executed by a processor, implements the multi-trajectory mapping method of any one of the various embodiments of this application.

[0029] In this embodiment, by aligning keyframe data corresponding to different historical driving trajectories and performing feature association and fusion processing, the accuracy and completeness of map data can be improved, reducing inaccurate or missing map information due to insufficient data sources. By processing keyframe data and performing feature association on multiple historical driving trajectories of the target vehicle on a preset driving route, the positioning accuracy of the target vehicle during driving can be improved, positioning errors can be reduced, and navigation accuracy can be improved. Based on the feature association results, fusion processing is performed to generate target map data, which can effectively reduce the cost and time of map updates, reduce the need for manual intervention, improve the update speed and efficiency of map data, and provide more accurate and complete map information for autonomous driving systems, thereby improving the performance and stability of autonomous driving systems and enhancing vehicle driving safety and efficiency. Therefore, the multi-trajectory mapping method provided in this embodiment achieves the goal of efficiently and accurately generating target map data, thus realizing the technical effect of improving the mapping efficiency and accuracy in the offline mapping process, and solving the technical problems of low mapping efficiency and poor accuracy in related technologies during offline mapping.

[0030] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0031] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0032] Figure 1 is a flowchart of a multi-trajectory mapping method according to one embodiment of this application;

[0033] Figure 2 is a flowchart of another multi-trajectory mapping method according to one embodiment of the present application;

[0034] Figure 3 is a schematic diagram of a multi-trajectory mapping method according to one embodiment of this application;

[0035] Figure 4 is a structural block diagram of a multi-track mapping device according to one embodiment of the present application. Detailed Implementation

[0036] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0037] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0038] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0039] This application provides a method embodiment for multi-trajectory mapping. It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than that shown here.

[0040] Offline mapping is a process of modeling and creating maps of the environment using pre-collected sensor data and map information, without the need for real-time positioning and sensor data. Offline mapping improves positioning accuracy and stability by leveraging historical data and map information, even in the absence of real-time data, while reducing the reliance on real-time sensor data and saving energy and computing resources. Furthermore, offline mapping allows for pre-modeling and analysis of the environment, enabling the early detection of potential hazards and obstacles, thus enhancing safety. In large-scale applications, it reduces the need for real-time data transmission and processing, lowering costs and complexity.

[0041] Traditional offline mapping methods based on single-shot route memory suffer from issues such as missed detections and false detections of mapping elements. Repeatedly collecting information from the same route for mapping can alleviate these problems. However, when using information collected multiple times for offline mapping, the time, road, and observed area differ each time, and issues such as unstable sensing effects and positioning signals also arise. This makes it difficult to accurately calculate the correlation between multiple collections of information, further affecting the efficiency and accuracy of offline mapping.

[0042] This application provides a multi-track mapping method, primarily applied to offline mapping processes in the field of autonomous driving technology. By efficiently and accurately generating map data, it provides crucial foundational data support for applications such as positioning and navigation of autonomous vehicles. This application can be applied to the creation and updating of maps for commuting routes, tourist routes, or other regular driving routes to adapt to constantly changing road environments and traffic conditions.

[0043] The multi-trajectory mapping method in this application effectively solves the problems of missed and false detections of mapping elements in traditional offline mapping methods, improving the accuracy of map data and the reliability of its application. By analyzing the historical driving trajectory of the target vehicle on a preset driving route, extracting keyframe data, and performing alignment and feature association processing, high-precision map data integrating multi-source information is finally generated. This high-precision map data not only improves the safety and efficiency of autonomous vehicles but also provides important road information resources for traffic management departments.

[0044] The multi-track mapping method provided in this application can be mainly applied to scenarios such as autonomous driving, intelligent transportation systems, vehicle navigation, and traffic monitoring. Its core advantage lies in enhancing the technical application and service quality in related technical scenarios by improving the efficiency and accuracy of map data generation.

[0045] In one implementation, in the offline map generation scenario for autonomous vehicles, these vehicles require accurate map data to assist navigation and decision-making. Historical driving trajectories can generate more precise offline maps. In the infrastructure planning scenario of intelligent transportation systems, accurate offline map data allows users to better understand and plan urban transportation infrastructure. In the updating and maintenance scenario of in-vehicle navigation systems, the vehicle's navigation system can provide more accurate navigation services by efficiently updating map data. More accurate map data helps the driver assistance system better understand road conditions, assists in precise vehicle positioning and efficient route planning, reduces travel time and costs, and improves driving safety.

[0046] Figure 1 is a flowchart of a multi-trajectory mapping method according to one embodiment of this application. As shown in Figure 1, the method includes the following steps:

[0047] Step S11: Obtain key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route. The key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0048] Step S12: Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0049] Step S13: Perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0050] Step S14: Perform fusion processing based on feature association results to generate target map data.

[0051] The aforementioned preset driving route can be a commuter route, a tourist route, or other regular driving routes. Multiple historical driving trajectories are trajectory data of the target vehicle driving multiple times on the preset driving route, which can be obtained, for example, through onboard sensors, GPS trajectory recorders, traffic monitoring cameras, and other devices. For each historical driving trajectory, trajectory analysis algorithms can be used to extract keyframe data. For example, the trajectory data can be processed through filtering, interpolation, and other methods to extract the vehicle's current attitude at key moments and all observable map features at that current attitude.

[0052] The keyframe data mentioned above may include, but is not limited to, timestamps, vehicle pose information, at least one map feature, route information, and environmental information. The timestamp records the acquisition time of the vehicle pose information and map features. Vehicle pose information includes the target vehicle's position and attitude, typically GPS data and Inertial Measurement Unit (IMU) data, which can be acquired through sensors on the target vehicle. At least one map feature can be a landmark such as a road sign, intersection, or building, or traffic facilities such as lane markings or traffic lights. Route information records the specific route taken by the target vehicle, including the origin, destination, and points of interest along the way. Environmental information may include surrounding traffic conditions and road conditions, which can help analyze the reasons for historical driving trajectories and decision-making processes.

[0053] After acquiring keyframe data, the keyframe data corresponding to different historical driving trajectories are aligned to obtain multiple keyframe pairing results. For example, this can be achieved by calculating the similarity between keyframes on different historical driving trajectories. First, the similarity between each pair of keyframes needs to be calculated, which can be done using a feature matching algorithm or a deep learning model. Then, the keyframes are sorted according to the similarity results, and the keyframe with the highest similarity is selected for pairing. Finally, multiple keyframe pairing results can be obtained, thus achieving the alignment of different historical driving trajectories. For example, to align the keyframe data corresponding to historical driving trajectory A with the keyframe data corresponding to historical driving trajectory B, a keyframe A1 on historical driving trajectory A can be matched with a keyframe B2 on historical driving trajectory B, thus forming a keyframe pair.

[0054] Furthermore, feature association processing is performed on multiple keyframe pairing results to obtain feature association results. For example, feature association is performed between each keyframe pairing result to construct constraints between keyframes. Feature association processing can associate and align map features between keyframe pairs, thereby determining the feature correspondence between multiple keyframe pairing results. Exemplarily, feature association processing includes two steps: coarse association and fine association. The coarse association process uses only lane line information from map features for association, while the fine association process uses all map features for association.

[0055] After obtaining the feature association results, a fusion process is performed based on these results to generate target map data. This target map data is an offline mapping result, such as a 2D map, a 3D map, or other forms of map data, providing accurate foundational data support for subsequent applications such as positioning and navigation. Fusing the feature association results to generate target map data provides more accurate and comprehensive map information, thereby further meeting users' offline mapping needs.

[0056] Based on steps S11 to S14 above, by aligning keyframe data corresponding to different historical driving trajectories and performing feature association and fusion processing, the accuracy and completeness of map data can be improved, reducing inaccurate or missing map information due to insufficient data sources. By processing keyframe data and performing feature association on multiple historical driving trajectories of the target vehicle on a preset driving route, the positioning accuracy of the target vehicle during driving can be improved, positioning errors can be reduced, and navigation accuracy can be improved. Finally, based on the feature association results, fusion processing is performed to generate target map data, which can effectively reduce the cost and time of map updates, reduce the need for manual intervention, improve the update speed and efficiency of map data, and provide more accurate and complete map information for autonomous driving systems, thereby improving the performance and stability of autonomous driving systems and enhancing vehicle driving safety and efficiency. Therefore, the multi-trajectory mapping method provided in this application embodiment achieves the goal of efficiently and accurately generating target map data using keyframe data collected from multiple historical driving trajectories, thereby achieving the technical effect of improving the mapping efficiency and accuracy in the offline mapping process, and solving the technical problems of low mapping efficiency and poor accuracy in related technologies when performing offline mapping.

[0057] The multi-trajectory mapping method in the embodiments of this application will be further described below.

[0058] In an optional embodiment, the multiple historical driving trajectories include at least: a first driving trajectory and a second driving trajectory, and the key frame data includes at least: multiple first key frames corresponding to the first driving trajectory and multiple second key frames corresponding to the second driving trajectory. In step S12, the key frame data corresponding to different historical driving trajectories are aligned to obtain multiple key frame pairing results, including:

[0059] Step S121: Perform position association processing on multiple first keyframes and multiple second keyframes to obtain position association results, wherein the position association results are used to represent the initial pairing results of the first keyframes and the second keyframes;

[0060] Step S122: Perform spatiotemporal association processing based on the location association results to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent the keyframe chain that satisfies the preset spatiotemporal continuity condition.

[0061] Step S123: Determine the pairing results of multiple keyframes based on the spatiotemporal correlation results.

[0062] For example, taking historical driving trajectory A as the first driving trajectory and historical driving trajectory B as the second driving trajectory, the multiple first keyframes corresponding to historical driving trajectory A are A1, A2, A3, and A4, and the multiple second keyframes corresponding to historical driving trajectory B are B1, B2, B3, B4, B5, and B6. By performing position association processing on the multiple first keyframes and multiple second keyframes, the position association result can be obtained, thereby enabling the acquisition of the second keyframe associated with each first keyframe.

[0063] Furthermore, spatiotemporal correlation processing is performed based on the location correlation results to obtain the spatiotemporal correlation result, which is a keyframe chain that meets preset spatiotemporal continuity conditions. For example, in the location correlation result, the second keyframe B2 is most strongly correlated with the first keyframe A1, and the second keyframe B3 is most strongly correlated with the first keyframe A2. During the alignment process of keyframe data corresponding to different historical driving trajectories, the correlation of keyframes is based on spatiotemporal consistency and continuity. That is, if adjacent first keyframes A1 and A2 in historical driving trajectory A are continuous in time and space, then the second keyframes B2 and B3 on their corresponding historical driving trajectory B also need to be continuous in time and space. According to spatiotemporal continuity, the longest continuous keyframe chain that meets the preset spatiotemporal continuity conditions is the final spatiotemporal correlation result. That is, the deviation between the temporal and spatial distances between A1 and A2 and between B2 and B3 is within a specified range. After obtaining the keyframe chain that satisfies the preset spatiotemporal continuity condition in the first and second driving trajectories, the final pairing result of multiple keyframes can be determined based on the keyframe chain. For example, the pairing result of multiple keyframes can be (A1-B2), (A2-B3), (A3-B4), (A4-B6).

[0064] Based on the above optional embodiments, by performing position association processing on multiple first keyframes and multiple second keyframes, a position association result is obtained. Then, based on the position association result, spatiotemporal association processing is performed to obtain a spatiotemporal association result. Finally, based on the spatiotemporal association result, the pairing result of multiple keyframes is determined. This allows for the consideration of temporal and spatial continuity when associating multiple trajectory information, thereby accurately calculating the association relationship between multiple collected information and further improving the mapping efficiency and accuracy in the offline mapping process.

[0065] In an optional embodiment, in step S13, feature association processing is performed on the pairing results of multiple keyframes to obtain feature association results including:

[0066] Step S131: Obtain vehicle pose information and lane line information based on the pairing results of multiple keyframes;

[0067] Step S132: Perform the first association process using vehicle pose information and lane line information to obtain lane line association results, wherein the lane line association results are used to represent the lane line correspondence in the keyframe pairing results.

[0068] Step S133: Perform a second association process using the lane line association results to obtain the feature association results.

[0069] For example, continuing with the example of multiple keyframe pairing results (A1-B2), (A2-B3), (A3-B4), (A4-B6), the vehicle pose information and lane line information of each keyframe are obtained based on the multiple keyframe pairing results. Among them, four lane lines a1, a2, a3, and a4 can be observed in the vehicle pose information corresponding to A1, and three lane lines b1, b2, and b3 can be observed in the vehicle pose information corresponding to B2.

[0070] Furthermore, the vehicle pose information and lane line information are used for the first association process to obtain the lane line association result. The first association process is a coarse association process, in which lane lines and lane boundaries need to be associated separately. During feature association, the spatial continuity of adjacent features needs to be guaranteed. If the positional deviation of the associated features between adjacent keyframes is greater than a preset value, then the adjacent keyframes are determined to be spatially discontinuous; if the positional deviation is less than or equal to the preset value, then the adjacent keyframes are determined to be spatially continuous. The optimal feature chain that is continuous in time and space is the final lane line association result, that is, after coarse processing, the correspondence between (a1-b1), (a2-b2), and (a3-b4) can be output.

[0071] After obtaining the lane line association results, the lane line association results are used for the second association processing, which is a fine association processing process. After the fine association processing, the feature association results are obtained.

[0072] Based on the above optional embodiments, vehicle pose information and lane line information are obtained based on the pairing results of multiple keyframes. Then, the vehicle pose information and lane line information are used for the first association processing to obtain the lane line association result. Finally, the lane line association result is used for the second association processing, which can quickly obtain accurate feature association results, and then be used for accurate offline mapping, thereby improving mapping efficiency and accuracy.

[0073] In an optional embodiment, in step S132, the first association process is performed using the vehicle pose information and lane line information to obtain the lane line association result, including:

[0074] Step S1321: Use lane line information to perform feature matching to obtain the first matching result;

[0075] Step S1322: The first matching result is processed based on the vehicle pose information to obtain the target detection result. The target detection result is used to determine whether the lane line association distance between adjacent keyframes meets the preset distance condition.

[0076] Step S1323: Determine the lane line association result based on the target detection result.

[0077] For example, in the coarse association process, lane line information is used for feature matching, which can associate lane lines and lane boundaries in keyframes to obtain the first matching result. Further, during feature association, the spatial continuity of adjacent features needs to be ensured. Based on vehicle pose information, the first matching result is processed for detection to obtain the target detection result. For example, based on the first matching result, the lane lines of the first keyframe A1 and the second keyframe B2 are associated. The adjacent first keyframe A2 and the second keyframe B3 are also associated. To ensure the temporal and spatial continuity of the association, the first matching result is processed for detection based on vehicle pose information. If the lane line association distance between A1 and B2 is consistent with the lane line association distance between A2 and B3, that is, the lane line association distance between adjacent keyframes meets the preset distance condition, then the lane line association result can be determined based on the target detection result. If only the lane line association between A1 and B2 is considered in the coarse association, pairing errors may occur. By combining the target detection results of multiple consecutive keyframes to determine the lane line association result, temporal and spatial correlation can be effectively considered, thereby effectively reducing association errors and improving the accuracy of offline mapping.

[0078] Based on the above optional embodiments, by using lane line information for feature matching, a first matching result is obtained. Then, based on vehicle pose information, the first matching result is processed to obtain a target detection result. Finally, based on the target detection result, the lane line association result is determined. This can improve the accuracy of the association relationship between multiple collected information and further improve the mapping efficiency and accuracy of offline mapping.

[0079] In an optional embodiment, in step S133, a second association process is performed using the lane line association results to obtain feature association results including:

[0080] Step S1331: Adjust the vehicle pose information based on the lane line association results to obtain the pose adjustment result;

[0081] Step S1332: Perform a second association process based on at least one map feature from the pose adjustment result and the keyframe pairing result to obtain the feature association result.

[0082] For example, the lane line association result includes the lane correspondences output after coarse processing, namely (a1-b1), (a2-b2), and (a3-b4). Based on the lane line association result, the vehicle pose information of the first keyframe A1 and the second keyframe B2 is adjusted so that the lane lines are aligned and overlapped in space, resulting in a pose adjustment result. Fine association processing is then performed based on at least one map feature from the pose adjustment result and the keyframe pairing result to obtain the feature association result. By adjusting the pose of the keyframes and the positions of the map features, all map features across multiple trajectories can be aligned together, thus enabling the fine association processing and improving the accuracy of the feature association result.

[0083] In an optional embodiment, in step S1332, the second association processing is performed based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain the feature association result, including: performing feature matching based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a second matching result; performing pose update processing based on the second matching result to obtain a pose update result; and determining the feature association result based on the pose update result in response to the pose update result satisfying a preset convergence condition.

[0084] For example, during feature matching, the discrete features to be matched include, but are not limited to, lane lines, sidewalks, stop lines, arrows, and road edges. After performing feature matching based on at least one map feature from the pose adjustment result and keyframe pairing result, a second matching result is obtained. This second matching result contains the matching relationships between the discrete features, and nonlinear optimization can be performed based on these relationships. The above feature matching process and pose update process are repeated until a preset convergence condition is met, i.e., the optimal matching result is reached, thus obtaining the final feature association result.

[0085] The goal of the aforementioned nonlinear optimization is to minimize the distance between matching pairs of lane lines and ground features, while also satisfying the combined inertial navigation constraints between two consecutive keyframes. Finally, the optimized pose update result for each keyframe is output, ensuring that the lane lines of historical driving trajectories A and B are aligned under the new pose.

[0086] Based on the above optional embodiments, feature matching is performed on at least one map feature from the pose adjustment result and keyframe pairing result to obtain a second matching result. Then, pose update processing is performed based on the second matching result to obtain a pose update result. Finally, in response to the pose update result satisfying the preset convergence condition, feature association result is determined based on the pose update result. This can further improve the reliability of feature association result through fine association processing, thereby improving the accuracy of offline mapping.

[0087] In an optional embodiment, step S14, performing fusion processing based on the feature association results to generate target map data includes:

[0088] Step S141: The feature association results are filtered using preset filtering conditions to obtain the filtered results. The preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results.

[0089] Step S142: Perform fusion processing based on the filtering results to generate target map data.

[0090] For example, after obtaining the feature association results, an accurate error filtering mechanism is introduced. Preset filtering conditions are used to filter out incorrect pairings in the feature association results, effectively eliminating erroneous associations and ensuring the accuracy of the mapping. Unaligned road segments in the nonlinear optimization results are checked, and their associations are removed. Mapping is not performed on these road segments, ensuring that no erroneous maps are generated. Offline mapping based on the association information in the filtering results can generate more accurate and complete target map data. Using this target map data, users can more accurately locate and navigate, improving the usability of the target map data and the user experience.

[0091] Based on the above optional embodiments, by using preset filtering conditions to filter the feature association results, the resulting filtered data can be effectively excluded from the association data that does not meet the conditions, reducing errors in the data and further improving the accuracy and reliability of the final generated target map data. Furthermore, preset filtering conditions and automated fusion processing can reduce manual intervention, improve the efficiency and speed of offline mapping, and save time and costs.

[0092] Figure 2 is a flowchart of another multi-trajectory mapping method according to one embodiment of the present application. As shown in Figure 2, the method includes the following steps:

[0093] Step S201: Obtain key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route. The key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0094] Step S202: Perform position association processing on multiple first keyframes and multiple second keyframes to obtain position association results, wherein the position association results are used to represent the initial pairing results of the first keyframes and the second keyframes;

[0095] Step S203: Perform spatiotemporal association processing based on the location association results to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent the keyframe chain that satisfies the preset spatiotemporal continuity condition.

[0096] Step S204: Determine the pairing results of multiple keyframes based on the spatiotemporal correlation results;

[0097] Step S205: Obtain vehicle pose information and lane line information based on the pairing results of multiple keyframes;

[0098] Step S206: Perform the first association process using vehicle pose information and lane line information to obtain lane line association results, wherein the lane line association results are used to represent the lane line correspondence in the keyframe pairing results;

[0099] Step S207: Perform a second association process using the lane line association results to obtain the feature association results;

[0100] Step S208: The feature association results are filtered using preset filtering conditions to obtain the filtered results. The preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results.

[0101] Step S209: Perform fusion processing based on the filtering results to generate target map data.

[0102] Based on the above steps S201 to S209, key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route are acquired. Then, the key frame data corresponding to different historical driving trajectories are aligned to obtain multiple key frame pairing results. Subsequently, feature association processing is performed on the multiple key frame pairing results to obtain feature association results. Finally, fusion processing is performed based on the feature association results to generate target map data. This achieves the goal of efficiently and accurately generating target map data using key frame data collected from multiple historical driving trajectories, thereby improving the mapping efficiency and accuracy in the offline mapping process. This solves the technical problems of low mapping efficiency and poor accuracy in related technologies when performing offline mapping.

[0103] Figure 3 is a schematic diagram of a multi-trajectory mapping method according to one embodiment of this application. As shown in Figure 3, the target vehicle has multiple historical driving trajectories on a preset driving route, including historical driving trajectory A, historical driving trajectory B, historical driving trajectory C, etc. Keyframe data corresponding to each historical driving trajectory is obtained. The keyframe data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information. The keyframe data corresponding to different historical driving trajectories are aligned to obtain multiple keyframe pairing results. Further, vehicle pose information and lane line information are obtained based on the multiple keyframe pairing results. Coarse association processing is performed using the vehicle pose information and lane line information to obtain lane line association results. Fine association processing is performed using the lane line association results to obtain feature association results. The feature association results are filtered using preset filtering conditions. The preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results. The filtered results that pass the preset filtering conditions are fused and mapped to generate target map data. The feature association results that do not pass the preset filtering conditions are aligned again.

[0104] In this embodiment, by collecting data from multiple trips and building maps offline, the problem of insufficient single-trip mapping capability can be solved. Furthermore, the keyframe association method based on temporal and spatial continuity in this embodiment can effectively solve the erroneous association problem existing in related technologies based on spatial proximity. The temporal and spatial continuity feature association method in this embodiment can effectively improve the accuracy of feature matching. Moreover, by employing iterative feature optimization and an accurate erroneous association filtering mechanism, target map data can be generated efficiently and accurately using keyframe data collected from multiple historical driving trajectories.

[0105] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0106] This application also provides a multi-track mapping device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0107] Figure 4 is a structural block diagram of a multi-track mapping device according to one embodiment of this application. As shown in Figure 4, the device includes:

[0108] The acquisition module 401 is used to acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route. The key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0109] Processing module 402 is used to align key frame data corresponding to different historical driving trajectories to obtain multiple key frame pairing results;

[0110] The association module 403 is used to perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0111] The generation module 404 is used to perform fusion processing based on feature association results to generate target map data.

[0112] In one embodiment, the processing module 402 is further configured to: perform position association processing on a plurality of first keyframes and a plurality of second keyframes to obtain position association results, wherein the position association results are used to represent the initial pairing results of the first keyframes and the second keyframes; perform spatiotemporal association processing based on the position association results to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent the keyframe chain that satisfies the preset spatiotemporal continuity condition; and determine the pairing results of a plurality of keyframes based on the spatiotemporal association results.

[0113] In one embodiment, the association module 403 is further configured to: obtain vehicle pose information and lane line information based on multiple keyframe pairing results; perform a first association process using the vehicle pose information and lane line information to obtain a lane line association result, wherein the lane line association result is used to represent the lane line correspondence in the keyframe pairing results; and perform a second association process using the lane line association result to obtain a feature association result.

[0114] In one embodiment, the association module 403 is further configured to: perform feature matching using lane line information to obtain a first matching result; perform detection processing on the first matching result based on vehicle pose information to obtain a target detection result, wherein the target detection result is used to determine whether the lane line association distance between adjacent keyframes meets a preset distance condition; and determine the lane line association result based on the target detection result.

[0115] In one embodiment, the association module 403 is further configured to: adjust the vehicle pose information based on the lane line association result to obtain a pose adjustment result; and perform a second association process based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a feature association result.

[0116] In one embodiment, the association module 403 is further configured to: perform feature matching based on at least one map feature in the pose adjustment result and the keyframe pairing result to obtain a second matching result; perform pose update processing based on the second matching result to obtain a pose update result; and determine the feature association result based on the pose update result in response to the pose update result satisfying a preset convergence condition.

[0117] In one embodiment, the generation module 404 is further configured to: filter the feature association results using preset filtering conditions to obtain filtering results, wherein the preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results; and perform fusion processing based on the filtering results to generate target map data.

[0118] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0119] In one embodiment, the processor described above can be configured to perform the following steps via a computer program:

[0120] S1, acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0121] S2, Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0122] S3, perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0123] S4 performs fusion processing based on feature association results to generate target map data.

[0124] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0125] In one embodiment, the computer-readable storage medium described above may be configured to store a computer program for performing the following steps:

[0126] S1, acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0127] S2, Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0128] S3, perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0129] S4 performs fusion processing based on feature association results to generate target map data.

[0130] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0131] In one embodiment, the computer program included in the above-described computer program product is executed by a processor in the following steps:

[0132] S1, acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0133] S2, Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0134] S3, perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0135] S4 performs fusion processing based on feature association results to generate target map data.

[0136] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0137] In one embodiment, the computer program stored on the non-volatile computer-readable storage medium is executed by a processor in the following steps:

[0138] S1, acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0139] S2, Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0140] S3, perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0141] S4 performs fusion processing based on feature association results to generate target map data.

[0142] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0143] In one embodiment, the above-described computer program is executed by a processor in the following steps:

[0144] S1, acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information.

[0145] S2, Align the keyframe data corresponding to different historical driving trajectories to obtain multiple keyframe pairing results;

[0146] S3, perform feature association processing on multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence between multiple keyframe pairing results;

[0147] S4 performs fusion processing based on feature association results to generate target map data.

[0148] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0149] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0150] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0151] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0152] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0153] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0154] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0155] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A multi-trajectory mapping method, characterized in that, include: Acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information. Alignment processing is performed on the key frame data corresponding to different historical driving trajectories to obtain multiple key frame pairing results; The multiple keyframe pairing results are subjected to feature association processing to obtain feature association results, wherein the feature association results are used to determine the feature correspondence relationship between the multiple keyframe pairing results; Based on the feature association results, a fusion process is performed to generate target map data.

2. The multi-trajectory mapping method according to claim 1, characterized in that, The plurality of historical driving trajectories includes at least: a first driving trajectory and a second driving trajectory, and the keyframe data includes at least: a plurality of first keyframes corresponding to the first driving trajectory and a plurality of second keyframes corresponding to the second driving trajectory. The alignment processing of the keyframe data corresponding to the different historical driving trajectories to obtain the pairing results of the plurality of keyframes includes: Position association processing is performed on the plurality of first keyframes and the plurality of second keyframes to obtain a position association result, wherein the position association result is used to represent the initial pairing result of the first keyframes and the second keyframes; Based on the location association results, spatiotemporal association processing is performed to obtain spatiotemporal association results, wherein the spatiotemporal association results are used to represent keyframe chains that satisfy preset spatiotemporal continuity conditions. The pairing results of the multiple keyframes are determined based on the spatiotemporal correlation results.

3. The multi-trajectory mapping method according to claim 1, characterized in that, The feature association processing of the multiple keyframe pairing results to obtain the feature association results includes: The vehicle pose information and lane line information are obtained based on the pairing results of the multiple keyframes. The vehicle pose information and the lane line information are used to perform a first association process to obtain a lane line association result, wherein the lane line association result is used to represent the lane line correspondence in the keyframe pairing result; The lane line association results are used to perform a second association process to obtain the feature association results.

4. The multi-trajectory mapping method according to claim 3, characterized in that, The first association process, which utilizes the vehicle pose information and the lane line information to obtain the lane line association result, includes: The lane line information is used to perform feature matching to obtain a first matching result; The first matching result is processed based on the vehicle pose information to obtain a target detection result, wherein the target detection result is used to determine whether the lane line association distance between adjacent keyframes meets a preset distance condition. The lane line association result is determined based on the target detection result.

5. The multi-trajectory mapping method according to claim 3, characterized in that, The second association process, which utilizes the lane line association results to obtain the feature association results, includes: The vehicle pose information is adjusted based on the lane line association results to obtain the pose adjustment result. A second association process is performed on at least one map feature from the pose adjustment result and the keyframe pairing result to obtain the feature association result.

6. The multi-trajectory mapping method according to claim 5, characterized in that, The second association process, based on at least one map feature from the pose adjustment result and the keyframe pairing result, yields the feature association result, including: Based on at least one map feature in the pose adjustment result and the keyframe pairing result, feature matching is performed to obtain a second matching result; Based on the second matching result, a pose update process is performed to obtain the pose update result; In response to the pose update result satisfying a preset convergence condition, the feature association result is determined based on the pose update result.

7. The multi-trajectory mapping method according to claim 1, characterized in that, The process of fusing the feature association results to generate the target map data includes: The feature association results are filtered using preset filtering conditions to obtain the filtered results, wherein the preset filtering conditions are used to filter out incorrect pairing relationships in the feature association results; The target map data is generated by performing a fusion process based on the filtering results.

8. The multi-trajectory mapping method according to claim 1, characterized in that: The keyframe data is extracted from the historical driving trajectory using a trajectory analysis algorithm.

9. The multi-trajectory mapping method according to claim 1, characterized in that, The alignment process of the keyframe data corresponding to different historical driving trajectories yields multiple keyframe pairing results, including: Based on the similarity between the keyframe data corresponding to different historical driving trajectories, the keyframe data corresponding to the different historical driving trajectories are aligned to obtain multiple keyframe pairing results.

10. The multi-trajectory mapping method according to claim 9, characterized in that, The step involves aligning the keyframe data corresponding to different historical driving trajectories based on the similarity between them, resulting in multiple keyframe pairing results, including: For the keyframe data corresponding to different historical driving trajectories, calculate the similarity between each pair of keyframes in the keyframe data; The keyframes are sorted according to the similarity results; Keyframes with high similarity are selected for pairing to obtain multiple keyframe pairing results.

11. The multi-trajectory mapping method according to claim 1, characterized in that: The feature association processing includes coarse association processing and fine association processing. Specifically, the coarse association process only uses lane line information from map features for association. The detailed association process uses all map features for association.

12. A multi-trajectory mapping device, characterized in that, include: The acquisition module is used to acquire key frame data corresponding to multiple historical driving trajectories of the target vehicle on a preset driving route, wherein the key frame data includes vehicle pose information in the historical driving trajectory and at least one map feature corresponding to the vehicle pose information. The processing module is used to align the key frame data corresponding to different historical driving trajectories to obtain multiple key frame pairing results; The association module is used to perform feature association processing on the multiple keyframe pairing results to obtain feature association results, wherein the feature association results are used to determine the feature correspondence relationship between the multiple keyframe pairing results; The generation module is used to perform fusion processing based on the feature association results to generate target map data.

13. A vehicle, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the multi-trajectory mapping method according to any one of claims 1 to 11.

14. A non-volatile storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the multi-trajectory mapping method according to any one of claims 1 to 11 when it is run.

15. A computer program product, characterized in that, The computer program product includes computer instructions that, when executed by a processor, implement the multi-trajectory mapping method as described in any one of claims 1 to 11.

Images