A parking lot data matching method, device, equipment, medium and vehicle

Abstract

The embodiment of the application discloses a parking lot data matching method, device, equipment, medium and vehicle, the method comprises the following steps: selecting vehicle driving tracks collected at different times from multiple times of collected parking lot driving tracks, taking one of the driving tracks as a current reference track and taking another driving track as a current to-be-matched track; based on aligned frames in which vehicle poses in the current reference track and the current to-be-matched track are matched, relative poses between each to-be-matched track frame, and an absolute position of the vehicle outside the parking lot, adjusting poses of each to-be-matched track frame in sequence, aligning the track frame after the pose adjustment with a corresponding reference track frame, and obtaining a current alignment result; returning to perform a matching operation of the current reference track and the current to-be-matched track until a predetermined number of parking lot driving track matching operations are completed. Through the above technical scheme, the multiple times of collected parking lot tracks are quickly and accurately associated.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of automatic driving, in particular, to a parking lot data matching method, device, equipment, medium and vehicle. BACKGROUND

[0002] At present, the functions of autonomous parking and automatic navigation of unmanned vehicles in parking lots are increasingly valued, and high-precision maps are needed to realize autonomous parking and automatic navigation in parking lots, so it is particularly important to accurately map the parking lot.

[0003] A large multi-story parking lot often needs to collect data multiple times to cover completely. The parking lot is often built underground, without GPS (Global Positioning System) signals, and the parking lot is multi-story, and the scenes between the layers are very similar, which brings great difficulty to the mapping of the parking lot.

[0004] In related technologies, for multiple times of collecting parking lot data, the trajectory data collected each time is usually associated by manual means. Since the data volume of the vehicle trajectory frames in the multiple times of collecting data is large, the efficiency and accuracy of the manual data association method are both low. SUMMARY

[0005] Embodiments of the present application provide a parking lot data matching method, device, equipment, medium and vehicle to quickly and accurately associate multiple times of collecting parking lot trajectories.

[0006] The specific technical solutions are as follows:

[0007] In a first aspect, the embodiments of the present application provide a parking lot data matching method, comprising:

[0008] selecting vehicle driving trajectories collected at different times from multiple times of collecting parking lot driving trajectories, and taking one of the driving trajectories as a current reference trajectory and taking another driving trajectory as a current to-be-matched trajectory, wherein each driving trajectory includes a pre-parking lot trajectory, an in-parking lot trajectory and a post-parking lot trajectory;

[0009] matching the current reference trajectory and the current to-be-matched trajectory, in the matching process, based on the aligned frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory are matched, the relative poses between the to-be-matched trajectory frames in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, adjusting the poses of the to-be-matched trajectory frames in turn, and aligning the trajectory frames after the pose adjustment with the corresponding reference trajectory frames in the current reference trajectory to obtain a current alignment result corresponding to the current matching process;

[0010] the current alignment result as a new current reference trajectory, and selecting another unmatched driving trajectory from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and returning to perform a matching operation of the new current reference trajectory and the new current to-be-matched trajectory until a predetermined number of parking lot driving trajectory matching operations are completed, to obtain a target matching result of the vehicle pose.

[0011] According to the above scheme, in the process of matching the reference trajectories collected at different times with the to-be-matched trajectories, based on the alignment frames in which the vehicle poses in the reference trajectories and the to-be-matched trajectories match, the relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering and after leaving the parking lot, each to-be-matched trajectory frame can be sequentially adjusted in pose, and the trajectory frame adjusted in pose is aligned in pose with the position corresponding reference trajectory frame, so that the multiple collected driving trajectories can be accurately and quickly associated together. In the association process, by using the progressive pose adjustment method for each to-be-matched trajectory frame, the vehicle pose alignment can be more accurate.

[0012] Optionally, based on the alignment frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory match, the relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering and after leaving the parking lot, each to-be-matched trajectory frame is sequentially adjusted in pose, and the trajectory frame adjusted in pose is aligned in pose with the corresponding reference trajectory frame in the current reference trajectory, to obtain a current alignment result corresponding to the current matching process, including:

[0013] Each to-be-matched trajectory frame in the current to-be-matched trajectory is sequentially traversed, and for each current to-be-matched trajectory frame, a current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame is determined in the current reference trajectory;

[0014] The current reference trajectory frame and the current to-be-matched trajectory frame are matched;

[0015] If the matching is successful, a first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame is determined;

[0016] Based on the first relative pose, the absolute position information of the vehicle before entering and after leaving the parking lot in each driving trajectory, and the second relative poses between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, all to-be-matched trajectory frames that have been successfully matched are optimized in pose, and the to-be-matched trajectory frame optimized in pose and the reference trajectory frame corresponding to the position that has been successfully matched are taken as alignment frames;

[0017] Based on the current alignment frame, the pose of a to-be-matched trajectory frame adjacent to the current alignment frame is adjusted, and the pose-adjusted adjacent trajectory frame is taken as a new current to-be-matched trajectory frame, and the matching operation of the new current to-be-matched trajectory frame and the current reference trajectory frame corresponding to the position is performed until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed, and a current alignment result of the vehicle pose in the current matching process is obtained.

[0018] Optionally, after determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, the method provided by the embodiment of the application further includes:

[0019] Based on the first relative pose, absolute position information of the vehicle in each driving trajectory before entering and after leaving the parking lot, and second relative poses between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, pose optimization is performed on all reference trajectory frames in the current reference trajectory, and the pose-optimized to-be-matched trajectory frame and the pose-optimized reference trajectory frame matched successfully are taken as alignment frames.

[0020] It can be known from the above technical solution that when the topological graph corresponding to the current to-be-matched trajectory frame is optimized, pose optimization can be simultaneously performed on all reference trajectory frames in the current reference trajectory, so that more accurate pose alignment between the reference trajectory and the to-be-matched trajectory is realized.

[0021] Optionally, the matching of the current reference trajectory frame and the current to-be-matched trajectory frame includes:

[0022] The local feature information of the parking lot includes traffic signs, wall identification, road surface identification, and lane line identification.

[0023] If the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully; or

[0024] If the local feature information of the parking lot corresponding to the current reference trajectory frame is inconsistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully.

[0025] Optionally, the method provided by the embodiment of the application further includes:

[0026] If the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully, the current to-be-matched trajectory frame matched unsuccessfully is temporarily stored.

[0027] After obtaining the current alignment result, the temporarily-stored target trajectory frames with failed matching are sequentially traversed, and for any target trajectory frame with failed matching, the matching operation of the target trajectory frame with the pose optimized and the reference trajectory frame corresponding to the position is repeatedly performed based on the pose of the target trajectory frame optimized, until all the temporarily-stored target trajectory frames with failed matching are traversed, and a new current alignment result is obtained.

[0028] Correspondingly, the driving trajectory corresponding to the current alignment result is taken as a new current reference trajectory, another un-matched driving trajectory is selected from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and the matching operation of the new current reference trajectory and the new current to-be-matched trajectory is performed again, including:

[0029] The driving trajectory corresponding to the new current alignment result is taken as a new current reference trajectory, another un-matched driving trajectory is selected from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and the matching operation of the new current reference trajectory and the new current to-be-matched trajectory is performed again.

[0030] It can be known from the above technical solution that, by re-querying and matching the reference trajectory frame corresponding to the position of the target trajectory frame with failed matching, the alignment result of the vehicle pose can be more accurate.

[0031] Optionally, the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame is determined in the current reference trajectory, including:

[0032] In the case that the current to-be-matched trajectory frame belongs to the pre-parking lot trajectory or the post-parking lot trajectory:

[0033] The absolute position corresponding to the current to-be-matched trajectory frame is obtained, and a search is performed in the current reference trajectory within a first set distance range from the absolute position to obtain the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame;

[0034] Or, in the case that the current to-be-matched trajectory frame belongs to the in-parking lot trajectory:

[0035] According to the current position information corresponding to the current to-be-matched trajectory frame, a search is performed in the current reference trajectory within a first set distance range from the current position information to obtain a candidate reference trajectory frame;

[0036] It is judged whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same layer of the parking lot;

[0037] If they belong to the same layer, the candidate reference trajectory frame is taken as the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame.

[0038] Optionally, the determining whether the candidate reference track frame and the current track frame to be matched belong to the same floor of the parking lot comprises:

[0039] determining a shortest topological path from the current track frame to be matched to the candidate reference track frame;

[0040] detecting whether there is a slope road section in the shortest topological path;

[0041] if there is no slope road section, determining that the candidate reference track frame and the current track frame to be matched belong to the same floor of the parking lot; or,

[0042] if there is a slope road section, and if a difference between height information of an uphill road section and height information of a downhill road section in the slope road section is within a second set distance range, determining that the candidate reference track frame and the current track frame to be matched belong to the same floor of the parking lot; or,

[0043] if the difference between the height information of the uphill road section and the height information of the downhill road section in the slope road section exceeds the second set distance range, determining that the candidate reference track frame and the current track frame to be matched do not belong to the same floor of the parking lot.

[0044] According to the above technical solution, by determining the shortest topological path from the current track frame to be matched to the candidate reference track frame, the complex situation between various paths from the current track frame to be matched to the candidate reference track frame is avoided. By detecting the slope road section in the shortest topological path, and according to the height information of the uphill road section and the downhill road section in the slope road section, it can be effectively determined whether the candidate reference track frame in the current reference track and the current track frame to be matched belong to the same floor of the parking lot, so as to avoid the situation that the track frames not belonging to the same floor of the parking lot are mismatched, and the problem of cross-layer data error association is solved.

[0045] Optionally, the detecting whether there is a slope road section in the shortest topological path comprises:

[0046] determining whether a pitch angle of the vehicle relative to a horizontal plane is greater than a preset angle;

[0047] regarding a road section corresponding to a track frame whose pitch angle is greater than or equal to the preset angle as a slope road section, and regarding a road section corresponding to a track frame whose pitch angle is less than the preset angle as a non-slope road section.

[0048] Optionally, based on the first relative pose, absolute position information of the vehicle in each driving track before entering the parking lot and after leaving the parking lot, and the second relative pose between each track frame to be matched in the current driving track to be matched, pose optimization is performed on all successfully matched track frames to be matched, comprising:

[0049] constructing a first topological graph corresponding to the current to-be-matched trajectories and a second topological graph corresponding to the current reference trajectories, wherein the nodes of the first topological graph are the to-be-matched trajectory frames, and the adjacent to-be-matched trajectory frames are connected by edges, and the nodes of the second topological graph are the reference trajectory frames, and the adjacent reference trajectory frames are connected by edges;

[0050] based on the first relative poses, the absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, and the second relative poses between the to-be-matched trajectory frames in the current to-be-matched driving trajectory, the first topological graph is optimized by a posegraph optimization method, and the to-be-matched trajectory frames after pose optimization and the matched reference trajectory frames are used as alignment frames, wherein the nodes of the alignment frames in the first topological graph and the second topological graph are connected by edges.

[0051] In the above technical solution, the topological graph corresponding to the current to-be-matched trajectory frames and the topological graph corresponding to the reference trajectories are optimized by using a posegraph optimization method, so that the vehicle pose can be accurately pose-matched, thereby accurately associating the reference trajectory with the to-be-matched trajectory, and solving the problem of insufficient data association.

[0052] Optionally, before matching the current reference trajectory with the current to-be-matched trajectory, the method provided by the embodiment of the present application further includes:

[0053] based on the absolute position information before entering the parking lot and after leaving the parking lot, the first topological graph and the second topological graph are optimized respectively to reduce the deviation between the current to-be-matched trajectory and the current reference trajectory in the absolute position.

[0054] Optionally, the vehicle pose corresponding to each reference trajectory frame or the vehicle pose corresponding to each to-be-matched trajectory frame is obtained by any one of the following sensor data, or is obtained by fusing multiple sensor data as follows:

[0055] inertial measurement unit (IMU) data, image data, radar point cloud data, and odometer data.

[0056] In a second aspect, the embodiment of the present application further provides a parking lot data matching device, which includes:

[0057] a driving trajectory selection module configured to select vehicle driving trajectories collected at different times from multiple times collected driving trajectories of parking lots at will, and take one of the driving trajectories as a current reference trajectory and take another driving trajectory as a current to-be-matched trajectory, wherein each driving trajectory includes a trajectory before entering the parking lot, a trajectory in the parking lot, and a trajectory after leaving the parking lot;

[0058] The matching module is configured to match the current reference trajectory with a current trajectory to be matched, in the matching process, based on an alignment frame in which the vehicle poses in the current reference trajectory and the current trajectory to be matched match, relative poses between each trajectory frame in the current trajectory to be matched, and absolute position information of the vehicle before entering and after leaving the parking lot, sequentially adjusting poses of each trajectory frame to be matched, and aligning the trajectory frame to be matched after the pose adjustment with a corresponding reference trajectory frame in the current reference trajectory to obtain a current alignment result corresponding to the current matching process;

[0059] The data association module is configured to take a driving trajectory corresponding to the current alignment result as a new current reference trajectory, select another un-matched driving trajectory from the multiple collected driving trajectories in the parking lot as a new current trajectory to be matched, and return to perform a matching operation of the new current reference trajectory and the new current trajectory to be matched until a predetermined number of matching operations of the driving trajectories in the parking lot are completed, to obtain a target matching result of the vehicle pose.

[0060] Optionally, the matching module comprises:

[0061] The current reference trajectory frame determination unit is configured to sequentially traverse each trajectory frame to be matched in the current trajectory to be matched, and for each current trajectory frame to be matched, determine a current reference trajectory frame corresponding to the position of the current trajectory frame to be matched in the current reference trajectory;

[0062] The matching unit is configured to match the current reference trajectory frame with the current trajectory frame to be matched;

[0063] The relative pose determination unit is configured to, if the current reference trajectory frame and the current trajectory frame to be matched match successfully, determine a first relative pose between the current trajectory frame to be matched and the corresponding current reference trajectory frame;

[0064] The first pose optimization unit is configured to, based on the first relative pose, absolute position information of the vehicle before entering and after leaving the parking lot in each driving trajectory, and second relative poses between each trajectory frame to be matched in the current trajectory to be matched, perform pose optimization on all successfully matched trajectory frames to be matched, and take the trajectory frame to be matched after the pose optimization and the reference trajectory frame corresponding to the position of the matching success as an alignment frame;

[0065] The pose alignment unit is configured to, based on the current alignment frame, adjust the pose of the trajectory frame to be matched adjacent to the current alignment frame, take the trajectory frame to be matched after the pose adjustment and adjacent to the current alignment frame as a new current trajectory frame to be matched, and return to perform a matching operation of the new current trajectory frame to be matched with the current reference trajectory frame corresponding to the position, until all trajectory frames to be matched in the current trajectory to be matched are traversed, to obtain a current alignment result of the vehicle pose in the current matching process.

[0066] Optionally, the device provided by the embodiment of the present application further comprises:

[0067] The second pose optimization unit is configured to, after determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, perform pose optimization on all reference trajectory frames in the current reference trajectory based on the first relative pose, absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, and second relative poses between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, and take the pose-optimized to-be-matched trajectory frame and the pose-optimized reference trajectory frame matched successfully as the aligned frames.

[0068] Optionally, the matching unit comprises:

[0069] The local feature information acquisition subunit is configured to acquire local feature information of the parking lot, wherein the local feature information comprises traffic sign, wall identification, road surface identification, and lane line identification.

[0070] The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame; or

[0071] The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is inconsistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame.

[0072] Optionally, the device provided by the embodiment of the present application further comprises:

[0073] The matching failure frame temporary storage module is configured to, if the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully, temporarily store the current to-be-matched trajectory frame matched unsuccessfully.

[0074] The re-matching module is configured to sequentially traverse the temporarily stored target trajectory frames matched unsuccessfully, and for any one of the target trajectory frames matched unsuccessfully, repeatedly perform the matching operation of the pose-optimized target trajectory frame and the reference trajectory frame corresponding in position based on the optimized pose of the target trajectory frame, until all the temporarily stored target trajectory frames matched unsuccessfully are traversed, and a new current alignment result is obtained.

[0075] Correspondingly, the data association module is specifically configured to:

[0076] The new current alignment result corresponds to a driving track as a new current reference track, and another unmatched driving track is selected from the multiple collected parking lot driving tracks as a new current to-be-matched track, and the matching operation of the new current reference track and the new current to-be-matched track is returned to be executed until a predetermined number of parking lot driving track matching operations are completed, and a target matching result of the vehicle pose is obtained.

[0077] Optionally, the current reference track frame determination unit comprises:

[0078] The first search subunit is configured to, in the case that the current to-be-matched track frame belongs to the pre-parking lot track or the post-parking lot track, acquire an absolute position corresponding to the current to-be-matched track frame, and search in the current reference track within a first set distance range from the absolute position to obtain a current reference track frame corresponding to the position of the current to-be-matched track frame.

[0079] Or,

[0080] The second search subunit is configured to, in the case that the current to-be-matched track frame belongs to the in-parking lot track, search in the current reference track within a first set distance range from the current position information corresponding to the current to-be-matched track frame to obtain a candidate reference track frame.

[0081] The same layer judgment subunit is configured to judge whether the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot.

[0082] The current reference track frame determination subunit is configured to, if the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot, take the candidate reference track frame as the current reference track frame corresponding to the position of the current to-be-matched track frame.

[0083] Optionally, the same layer judgment subunit comprises:

[0084] The shortest topological path determination component is configured to determine a shortest topological path from the current to-be-matched track frame to the candidate reference track frame.

[0085] The slope road section detection component is configured to detect whether there is a slope road section in the shortest topological path.

[0086] The same layer determination component is configured to, if there is no slope road section in the shortest topological path, determine that the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot.

[0087] Or, if there is a slope road segment in the shortest topological path, and if the difference between the height information of the uphill road segment and the height information of the downhill road segment in the slope road segment is within a second set distance range, it is determined that the candidate reference trajectory frame and the current trajectory frame to be matched belong to the same floor of the parking lot;

[0088] Or, if the difference between the height information of the uphill road segment and the height information of the downhill road segment in the slope road segment exceeds the second set distance range, it is determined that the candidate reference trajectory frame and the current trajectory frame to be matched do not belong to the same floor of the parking lot.

[0089] Optionally, the same floor determination component is specifically configured to:

[0090] determine whether the pitch angle of the vehicle relative to the horizontal plane is greater than a preset angle;

[0091] determine the road segment corresponding to the trajectory frame whose pitch angle is greater than or equal to the preset angle as a slope road segment, and determine the road segment corresponding to the trajectory frame whose pitch angle is less than the preset angle as a non-slope road segment.

[0092] Optionally, the first pose optimization unit is specifically configured to:

[0093] construct a first topological graph corresponding to the current trajectory to be matched, and a second topological graph corresponding to the current reference trajectory, wherein the nodes of the first topological graph are the trajectory frames to be matched, and the adjacent trajectory frames to be matched are connected by edges, and the nodes of the second topological graph are the reference trajectory frames, and the adjacent reference trajectory frames are connected by edges;

[0094] based on the first relative pose, the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot in each driving trajectory, and the second relative pose between each trajectory frame to be matched in the current trajectory to be matched, optimize the first topological graph by a pose graph optimization method, and take the trajectory frame to be matched after pose optimization and the matched reference trajectory frame as an alignment frame, wherein the nodes of the alignment frame in the first topological graph and the second topological graph are connected by edges.

[0095] Optionally, the apparatus provided by the embodiment further comprises:

[0096] An initial optimization module is configured to, before matching the current reference trajectory and the current trajectory to be matched, optimize the first topological graph and the second topological graph based on the absolute position information before entering the parking lot and after leaving the parking lot, so as to reduce the deviation between the current trajectory to be matched and the current reference trajectory in the absolute position.

[0097] Optionally, the vehicle pose corresponding to each reference trajectory frame or the vehicle pose corresponding to each trajectory frame to be matched is obtained by any one of the following sensor data, or is obtained by fusing multiple sensor data as follows:

[0098] Inertial Measurement Unit, IMU, data, image data, radar point cloud data, and odometry data.

[0099] In a third aspect, an electronic device is provided, which comprises:

[0100] one or more processors;

[0101] a memory device for storing one or more programs,

[0102] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for matching parking lot data as provided in any embodiment of the present application.

[0103] In a fourth aspect, a storage medium is provided, which stores a computer program, and the program, when executed by a processor, implements the method for matching parking lot data as provided in any embodiment of the present application.

[0104] In a fifth aspect, a vehicle is provided, which comprises the matching device for parking lot data as provided in any embodiment of the present application, or comprises the electronic device as provided in any embodiment of the present application.

[0105] In a sixth aspect, a computer program is provided, which comprises program instructions, and the program instructions, when executed by a computer, implement the method for matching parking lot data as provided in any embodiment of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0106] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0107] Figure 1a A flowchart of the method for matching parking lot data provided in embodiment one of the present application;

[0108] Figure 1b A topological graph of vehicle driving trajectory containing a one-floor underground parking lot provided in embodiment one of the present application;

[0109] Figure 1c A topological graph of vehicle driving trajectory containing a two-floor underground parking lot provided in embodiment one of the present application;

[0110] Figure 1dThe topology graph provided by the embodiment one of the present application for matching the reference trajectory and the to-be-matched trajectory after initial pose optimization;

[0111] Figure 1e The topology graph provided by the embodiment one of the present application for matching the reference trajectory and the to-be-matched trajectory after initial pose optimization;

[0112] Figure 1f The topology graph provided by the embodiment one of the present application for matching the reference trajectory and the to-be-matched trajectory after initial pose optimization;

[0113] Figure 2 The flow chart of the parking lot data matching method provided by the embodiment two of the present application;

[0114] Figure 3 The structural block diagram of the parking lot data matching device provided by the embodiment three of the present application;

[0115] Figure 4 The structural block diagram of the electronic device provided by the embodiment four of the present application;

[0116] Figure 5 The schematic diagram of the vehicle provided by the embodiment five of the present application. DETAILED DESCRIPTION

[0117] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0118] It should be noted that the terms “include” and “have” and any variations thereof in the embodiments of the present application and the drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed or optionally further includes other steps or units inherent to the process, method, product or device.

[0119] The embodiments of the present application disclose a parking lot data matching method, device, equipment, medium and vehicle. The following are described in detail respectively.

[0120] Embodiment one

[0121] Figure 1aA flowchart of a parking lot data matching method provided for Embodiment One of the present application is shown in FIG. 1. The method can be applied to a vehicle terminal such as a vehicle-mounted computer, an industrial personal computer (IPC), etc., or a server. Embodiments of the present application do not limit the application. The method provided by the present embodiment can be applied to a scenario in which multiple sets of collected multi-layer parking lot driving trajectories are associated. The method provided by the present embodiment can be executed by a parking lot data matching device, which can be implemented in software and / or hardware. As shown in FIG. 1, the method provided by the present embodiment specifically includes the following steps. Figure 1a

[0122] S110, any of the multiple sets of collected parking lot driving trajectories is selected as a current reference trajectory and another set of driving trajectories is selected as a current to-be-matched trajectory.

[0123] In the multiple sets of collected parking lot driving trajectories, the vehicle pose corresponding to each frame of trajectory can be collected by a vehicle-mounted sensor or obtained by fusing multiple types of sensor data. The vehicle-mounted sensor can include an inertial measurement unit (IMU), an image sensor, a radar, an odometer, etc.

[0124] In the present embodiment, each of the multiple sets of collected parking lot driving trajectories includes a driving trajectory before entering the parking lot, a driving trajectory in the parking lot, and a driving trajectory after exiting the parking lot. The vehicle has a GPS (Global Positioning System) signal before entering the parking lot and after exiting the parking lot. The three trajectories, i.e., the trajectory before entering the parking lot, the driving trajectory in the parking lot, and the trajectory after exiting the parking lot, are continuous trajectories. By applying a GPS constraint to the two trajectories before and after entering and exiting the parking lot and using the continuity of the three trajectories, the entire trajectory can achieve a positioning accuracy of a meter level.

[0125] In the present embodiment, matching the current reference trajectory and the current to-be-matched trajectory actually means matching the vehicle poses corresponding to each frame of trajectory in the current reference trajectory and the current to-be-matched trajectory, and aligning the vehicle poses of each frame of to-be-matched trajectory and the corresponding reference trajectory by optimizing the vehicle poses. To reduce the amount of calculation, multiple key frames can be selected from the current reference trajectory as reference trajectory frames, and multiple key frames can be selected from the current to-be-matched trajectory as to-be-matched trajectory frames. The selection of each key frame can be that a trajectory frame with a preset distance is selected as a key frame. The preset distance can be any distance in a range of 1-3 meters.

[0126] ​In the embodiment, in order to realize accurate alignment of the poses between the current reference trajectory and the current to-be-matched trajectory, the poses of each reference trajectory frame and each to-be-matched trajectory frame can be respectively optimized. Specifically, a first topological graph corresponding to the current to-be-matched trajectory and a second topological graph corresponding to the current reference trajectory can be constructed, wherein the nodes of the first topological graph are the to-be-matched trajectory frames, and the adjacent to-be-matched trajectory frames are connected by edges, and the nodes of the second topological graph are the reference trajectory frames, and the adjacent reference trajectory frames are connected by edges. Before matching the current reference trajectory and the current to-be-matched trajectory, based on the absolute position information before entering and after leaving the parking lot, the first topological graph and the second topological graph are respectively optimized, which can reduce the deviation between the current to-be-matched trajectory and the current reference trajectory in the absolute position.

[0127] Specifically, Figure 1b The topological graph of the vehicle driving trajectory provided by the first embodiment of the application includes a one-floor underground parking lot (B1), Figure 1c The topological graph of the vehicle driving trajectory provided by the first embodiment of the application includes a two-floor underground parking lot (B2). As shown in Figure 1b And 1c The nodes in each driving trajectory are vehicle driving trajectory frames, i.e. optimization variables. Adjacent nodes are connected by edges, and adjacent nodes are constrained by relative poses. The GPS signals generated by GPS measurement before entering and after leaving the parking lot are constrained. One of Figure 1b And Figure 1c The driving trajectory is taken as a to-be-matched trajectory, and the other driving trajectory is taken as a reference trajectory, Figure 1d The topological graph of the vehicle driving trajectory provided by the first embodiment of the application includes a one-floor underground parking lot (B1), Figure 1b And Figure 1c After the pose graph optimization of Figure 1d As shown in

[0128] S120, match the current reference trajectory with the current to-be-matched trajectory, in the matching process, based on the aligned frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory match, the relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, sequentially adjust the poses of each to-be-matched trajectory frame, and align the trajectory frame after the pose adjustment with the corresponding reference trajectory frame in the current reference trajectory to obtain a current alignment result corresponding to the current matching process.

[0129] In the embodiment, for each to-be-matched trajectory frame in the to-be-matched trajectory, the pose of each to-be-matched trajectory frame can be optimized in a progressive optimization manner. Specifically, steps (1)-(3) can be used to achieve this.

[0130] (1) sequentially traverse each to-be-matched trajectory frame in the current to-be-matched trajectory, and for each current to-be-matched trajectory frame, determine a current reference trajectory frame in the current reference trajectory corresponding to the position of the current to-be-matched trajectory frame.

[0131] The current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame refers to a reference trajectory frame that belongs to the same parking lot as the current to-be-matched trajectory frame and is close to the position of the current to-be-matched trajectory frame, i.e., the distance is within a first set distance range. The number of reference trajectory frames that meet the above conditions can be one or multiple.

[0132] In the embodiment, the current to-be-matched trajectory frame can belong to the trajectory before entering the parking lot or the trajectory after leaving the parking lot, or it can also belong to the trajectory in the parking lot.

[0133] In the case where the current to-be-matched trajectory frame belongs to the trajectory before entering the parking lot or the trajectory after leaving the parking lot, since the vehicle can determine the absolute position information of the vehicle based on the received GPS signal, the search for the reference trajectory frame corresponding to the position can be directly performed based on the absolute position information, and specifically, the following method can be used to achieve this.

[0134] Obtain the absolute position corresponding to the current to-be-matched trajectory frame, and search within a first set distance range from the absolute position in the current reference trajectory to obtain the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame. The first set distance range can be 8-12 meters.

[0135] Alternatively, in the case where the current to-be-matched trajectory frame belongs to the trajectory in the parking lot, the vehicle cannot receive the GPS signal or the received GPS signal is weak, and in this case, the following steps A-C can be used to determine the current reference trajectory frame corresponding to the position of the current trajectory frame.

[0136] A. According to the current position information corresponding to the current to-be-matched trajectory frame, search in the current reference trajectory within a first set distance range from the current position information to obtain a candidate reference trajectory frame.

[0137] B. Determine whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot. If they belong to the same floor, the candidate reference trajectory frame is taken as the current reference trajectory frame corresponding to the position of the current trajectory frame.

[0138] There are various ways to determine whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot. In this embodiment, the height difference between the uphill road section and the downhill road section in the shortest topological path is detected, and the candidate reference trajectory frame and the current to-be-matched trajectory frame are determined to belong to the same floor of the parking lot by judging the height difference. The determination can be realized by the following way:

[0139] Determine the shortest topological path from the current to-be-matched trajectory frame to the candidate reference trajectory frame. Detect whether there is a slope road section in the shortest topological path. If there is no slope road section, it is determined that the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot. Or, if there is a slope road section, and if the height information of the uphill road section and the height information of the downhill road section in the slope road section are within a second set distance range, it is determined that the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot. Or, if the height information of the uphill road section and the height information of the downhill road section in the slope road section exceed the second set distance range, it is determined that the candidate reference trajectory frame and the current to-be-matched trajectory frame do not belong to the same floor of the parking lot. The second set distance range can be 3-5 meters.

[0140] In this embodiment, whether it is a slope road section can be determined by the pitch angle of the vehicle relative to the horizontal plane. Specifically, the road section corresponding to the trajectory frame with a pitch angle greater than or equal to a preset angle is taken as a slope road section, and the road section corresponding to the trajectory frame with a pitch angle less than the preset angle is taken as a non-slope road section. The preset angle can be in the range of 5°-10°. The pitch angle of the vehicle relative to the horizontal plane can be determined by the IMU, or the attitude of the vehicle can be determined based on the data fused by multiple sensors, and then the pitch angle of the vehicle relative to the horizontal plane is determined. The multiple sensors can include IMU, odometer, image sensor, etc.

[0141] In this embodiment, the height information of the uphill road section or the height information of the downhill road section can be calculated according to the height difference between the vehicle pose at the top of the slope road section and the vehicle pose at the bottom of the slope road section.

[0142] In this embodiment, for the candidate reference trajectory frame belonging to the same parking lot as the current to-be-matched trajectory, it is taken as the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame. For the candidate reference trajectory frame not belonging to the same parking lot as the current to-be-matched trajectory, it is not matched with the current to-be-matched trajectory. In this case, the next new to-be-matched trajectory frame is traversed backward, and the operation of determining the current reference trajectory frame corresponding to the position of the new current to-be-matched trajectory frame is returned, that is, the operation of step (1) is returned.

[0143] For example, Figure 1e A topology graph for matching a to-be-matched trajectory with a reference trajectory is provided for the first embodiment of the present application. As shown in Figure 1e K2, K4, K8, K7 and K6 are reference trajectory frames in the current reference trajectory. K1, K3, K9, K10, K5 and K11 are to-be-matched trajectory frames in the current to-be-matched trajectory, wherein K1 and K2, K3 and K4 are aligned frames with aligned poses. In the progressive localization matching process, the vehicle trajectory will have a significant drift in absolute position and even produce a wrong layer after entering the parking lot interior and being unable to receive a GPS signal. As shown in Figure 1e K5 and K6 in FIG. 5, since the trajectory where K5 is located gradually drifts, its position is very close to K6, which will cause the wrong association of the two, and this situation will cause the entire mapping to fail. In order to avoid this problem, the shortest topological path between the two needs to be queried from the topology graph when initiating the matching of the two frames each time, for example, the shortest topological path path_0 between K5 and K6 is: K5-K10-K9-K3-K4-K8-K7-K6.

[0144] In the shortest topological path path_0, the uphill and downhill situations on the shortest topological path are queried according to the pose of the to-be-matched trajectory frame. As shown in Figure 1eIf K3-K9 is directly descending from the outside of the parking lot to B2, for example, descending 6m, and K4-K8 is descending from the outside of the parking lot to B1, for example, descending 3m, it can be determined that the current to-be-matched trajectory frame K5 and the candidate reference trajectory frame K6 are not in the same floor, and thus K5 and K6 do not need to be matched, thereby excluding the case of false matching. At this time, the next to-be-matched trajectory frame K11 of the current to-be-matched trajectory frame needs to be searched again for a candidate reference trajectory frame, and if the candidate reference trajectory frame K6 is searched, it is determined whether K11 and K6 belong to the same floor, that is, the up-and-down slope on the shortest topological path between K11 and K6 is calculated. For example, K4-K8 descends 3m, K3-K9 descends 6m, and K5-K11 ascends 3m, and thus it can be determined that K11 and K6 are in the same floor, and at this time, the reference trajectory frame K6 can be taken as the current reference trajectory frame corresponding to the position of the current to-be-matched key frame K11, and the following step (ii) can be continued, that is, K6 and K11 are matched using the local feature information of the parking lot, and after K6 and K11 become the aligned frames with matched poses, K6 and K11 are connected in the topological graph.

[0145] In this embodiment, by determining the shortest topological path from the current to-be-matched trajectory frame to the candidate reference trajectory frame, the complex situation between various paths between the current to-be-matched trajectory frame and the candidate reference trajectory frame is avoided. By detecting the slope road section in the shortest topological path and according to the height information of the uphill and downhill sections in the slope road section, it can be effectively determined whether the candidate reference trajectory frame in the current reference trajectory and the current to-be-matched trajectory frame belong to the same floor of the parking lot, so as to avoid the case of mismatching the trajectory frames not belonging to the same floor of the parking lot, and solve the problem of cross-layer data error association.

[0146] (ii) matching the current reference trajectory frame and the current to-be-matched trajectory frame, and if the matching is successful, determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame.

[0147] As can be understood by those skilled in the art, the vehicle trajectories obtained by different driving of the vehicle in the same floor of the parking lot will have a deviation in pose, but the local feature information of the parking lot corresponding to the position of the trajectory frames in different driving trajectories is fixed. The local feature information of the parking lot can include traffic signs, wall identification, road surface identification, lane line identification, etc. The local feature information of the parking lot can be obtained by a vehicle-mounted sensor (such as an image sensor, a radar, etc.).

[0148] Based on the above reasons, in the embodiment, the current reference trajectory frame and the current to-be-matched trajectory frame are matched based on the acquired local feature information of the parking lot, if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully, if the local feature information of the parking lot corresponding to the current reference trajectory frame is inconsistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully.

[0149] In the embodiment, for the current to-be-matched trajectory frame and the corresponding current reference trajectory frame which are matched successfully, the relative pose of the two is calculated to associate the two, for the current to-be-matched trajectory frame which is matched unsuccessfully, it is temporarily stored, and after the traversal of all to-be-matched trajectory frames is completed, the to-be-matched trajectory frame which is matched unsuccessfully is re-matched.

[0150] (Three) based on the first relative pose, the absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, the pose of all successfully matched to-be-matched trajectory frames is optimized, and the to-be-matched trajectory frame after pose optimization and the reference trajectory frame corresponding to the matched position are taken as the alignment frame.

[0151] In the embodiment, the pose of the vehicle is optimized by constructing the pose topology graph and optimizing the pose topology graph, so as to realize the pose alignment of the to-be-matched trajectory frame and the reference trajectory frame corresponding to the matched position. It can be realized by the following steps 1-2:

[0152] 1. Construct the first topology graph corresponding to the current to-be-matched trajectory, and the second topology graph corresponding to the current reference trajectory.

[0153] For example, the topology graph shown in FIG. 1 is taken as the topology graph corresponding to the current reference trajectory, and the topology graph shown in FIG. 2 is taken as the topology graph corresponding to the current to-be-matched trajectory. Figure 1b As shown in FIG. 1, the nodes of the second topology graph corresponding to the current reference trajectory are each reference trajectory frame, and the adjacent reference trajectory frames are connected by edges. Figure 1c As shown in FIG. 2, the nodes of the first topology graph corresponding to the current to-be-matched trajectory are each to-be-matched trajectory frame, and the adjacent to-be-matched trajectory frames are connected by edges. Figure 1b Figure 1c As shown in FIG. 2, the nodes of the first topology graph corresponding to the current to-be-matched trajectory are each to-be-matched trajectory frame, and the adjacent to-be-matched trajectory frames are connected by edges.

[0154] ​2. based on the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, the absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, the first topological graph is optimized by a posegraph optimization method, and the to-be-matched trajectory frame after pose optimization and the matched reference trajectory frame are used as alignment frames, wherein the alignment frames in the first topological graph and the second topological graph are connected by edges.

[0155] In the embodiment, the topological graph corresponding to the current to-be-matched trajectory frame is optimized by using a posegraph optimization method, so that the to-be-optimized variable represented by the node, i.e., the vehicle pose, can be optimized. Figure 1c Specifically, in the optimization process, if the current to-be-matched trajectory frame is the Nth to-be-matched trajectory frame, the to-be-matched trajectory frame for pose optimization is from the 1st to-be-matched trajectory frame to the current Nth to-be-matched trajectory frame, and the optimized positions corresponding to the 1st to Nth nodes are obtained, and the N to-be-matched trajectory frames after pose optimization and the matched reference trajectory frame are used as alignment frames. In the embodiment, by using the posegraph optimization method, the vehicle poses in the to-be-matched trajectory and the reference trajectory can be accurately aligned, so that the to-be-matched trajectory and the reference trajectory are accurately associated together, and the problem of insufficient data association is solved.

[0156] Further, in order to make the pose alignment of the current to-be-matched trajectory frame and the current reference trajectory frame more accurate, when optimizing the topological graph corresponding to the current to-be-matched trajectory frame, the pose of all reference trajectory frames in the current reference trajectory can be optimized based on the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, the absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, so as to realize more accurate pose alignment between the reference trajectory and the to-be-matched trajectory.

[0157] For example, Figure 1f Another topological graph for matching the to-be-matched trajectory and the reference trajectory is provided for the first embodiment of the application. Figure 1d Based on the above, Figure 1fA further optimization process of the vehicle pose is shown. For the first to-be-matched trajectory frame K1 in the current to-be-matched trajectory data2, the trajectory frame K1 is a trajectory frame before entering the parking lot. According to the absolute position information of the trajectory frame K1 before entering the parking lot, the reference trajectory frame K2 corresponding to the pose of the to-be-matched trajectory frame K1 within a first set distance range can be queried in the reference driving trajectory data1. The K1 and K2 are matched by using the corresponding parking lot local feature information of K1 and K2, if the matching is successful, the relative pose of the vehicle corresponding to the K1 trajectory frame and the K2 trajectory frame is calculated, and after the calculation of the relative pose is completed, by optimizing the first topological graph and the second topological graph, the more accurate pose configuration between K1 and K2 can be realized. The current to-be-matched trajectory frame K1 and the corresponding reference trajectory frame K2 after the pose optimization are the aligned frames. In the topological graph, the nodes corresponding to K1 and K2 are connected, so that the current to-be-matched trajectory data2 and the current reference trajectory data1 obtain the initial registration.

[0158] (Four) based on the current aligned frame, adjusting the pose of the to-be-matched trajectory frame adjacent to the current aligned frame, and taking the pose-adjusted adjacent trajectory frame as a new current to-be-matched trajectory frame, and returning to perform the matching operation of the new current to-be-matched trajectory frame and the current reference trajectory frame corresponding to the position, until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed, and the current alignment result of the vehicle pose in the current matching process is obtained.

[0159] Those skilled in the art can understand that, since the relative pose between the two adjacent trajectory frames in the same driving trajectory frame is fixed and unchanged, when the optimized position of the Nth to-be-matched trajectory frame currently traversed is obtained by topological graph optimization, the pose of the next to-be-matched trajectory frame adjacent to the Nth to-be-matched trajectory frame currently traversed, i.e. the N+1th frame, also needs to be adjusted accordingly, and the N+1th to-be-matched trajectory frame after the pose adjustment is taken as the new current to-be-matched trajectory frame, and the operation of step (one) is returned to perform, to obtain the current reference trajectory frame corresponding to the position of the new current to-be-matched trajectory frame, and continue to perform the matching operation of the new current to-be-matched trajectory frame and the current reference trajectory frame corresponding to the position in step (two), and then continue to perform steps (three) and (four), until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed, and the current alignment result of the vehicle pose in the current matching process is obtained.

[0160] Specifically, as Figure 1fAs shown, the to-be-matched trajectory frame K1 and the to-be-matched trajectory frame K2 have been aligned, at which time the pose of the to-be-matched trajectory frame K1 has been optimized. Since the relative pose between the to-be-matched trajectory frame K1 and the adjacent trajectory frame K3 is fixed, after the pose of the to-be-matched trajectory frame K1 is adjusted, the adjacent trajectory frame K3 (solid black circle) can update the pose to the black dashed circle according to the relative pose between K1 and K3, that is, the position of the adjacent trajectory frame K3 is adjusted. The trajectory frame K3 after the pose adjustment is taken as a new current to-be-matched trajectory frame, and the search is performed again according to the position to obtain the corresponding reference trajectory frame K4 in the reference trajectory, and the parking lot local features are used to match the trajectory frame K4 and the trajectory frame K3, and the relative pose is calculated after the matching is successful, so as to realize the association of the trajectory frame K4 and the trajectory frame K3. The above process is repeated for all subsequent to-be-matched trajectory frames until the processing of all to-be-matched frames is completed.

[0161] In S130, the driving trajectory corresponding to the current alignment result is taken as a new current reference trajectory, another un-matched driving trajectory is selected from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and the matching operation of the new current reference trajectory and the new current to-be-matched trajectory is performed again until the predetermined number of parking lot driving trajectory matching operations are completed, and the target matching result of the vehicle pose is obtained.

[0162] The predetermined number can be set according to the number of the actually collected parking lot driving trajectories, for example, can be set as the total number of the collected parking lot driving trajectories, which is not specifically limited in this embodiment.

[0163] In the process of matching the reference trajectories collected at different times with the to-be-matched trajectories, based on the aligned frames in which the vehicle poses in the reference trajectories and the to-be-matched trajectories are matched, the relative poses between the to-be-matched trajectory frames in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, the vehicle pose of each to-be-matched trajectory frame can be adjusted in sequence, and the trajectory frame after the pose adjustment is aligned with the position corresponding reference trajectory frame, so that the multiple collected driving trajectories can be accurately associated. In the association process, by using the progressive pose adjustment mode for each to-be-matched trajectory frame, the vehicle pose alignment can be more accurate. Compared with the way of manually associating the multiple collected parking lot data in the related art, the method provided in this embodiment improves the efficiency and accuracy of the parking lot data association.

[0164] Embodiment Two

[0165] Figure 2A flowchart of a parking lot data matching method provided for the second embodiment of the present application is shown in the figure. The embodiment is based on the above-mentioned embodiment and details the case where the current reference trajectory frame and the current to-be-matched trajectory frame fail to match. As shown in the figure, the method provided by the embodiment includes the following steps. Figure 2

[0166] S200, randomly selecting vehicle driving trajectories collected at different times from the multiple collected parking lot driving trajectories, taking one of the driving trajectories as a current reference trajectory and taking another driving trajectory as a current to-be-matched trajectory.

[0167] S210, matching the current reference trajectory and the current to-be-matched trajectory, in the matching process, sequentially traversing each to-be-matched trajectory frame in the current to-be-matched trajectory, for each current to-be-matched trajectory frame, determining a current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame in the current reference trajectory.

[0168] S220, matching the current reference trajectory frame and the current to-be-matched trajectory frame, determining whether the two frames match successfully, if yes, performing step S230; otherwise, performing step S240.

[0169] S230, determining a first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, and continuing to perform step S250.

[0170] S240, temporarily storing the current to-be-matched trajectory frame that fails to match, and continuing to perform step S270.

[0171] S250, based on the first relative pose, the absolute position information of the vehicle in each driving trajectory before entering the parking lot and after leaving the parking lot, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched trajectory, performing pose optimization on all to-be-matched trajectory frames that have successfully matched and all reference trajectory frames in the current reference trajectory, and taking the pose-optimized to-be-matched trajectory frames and the pose-optimized reference trajectory frames that have successfully matched as alignment frames, and continuing to perform step S260.

[0172] In the embodiment, when the pose graph optimization method is used for pose optimization, the pose optimization is performed on all to-be-matched trajectory frames that have successfully matched and all reference trajectory frames in the current reference trajectory, which can make the pose alignment of the current to-be-matched trajectory frame and the current reference trajectory frame more accurate.

[0173] ​S260, based on the current alignment frame, adjusting the pose of the to-be-matched trajectory frame adjacent to the current alignment frame, taking the pose-adjusted adjacent trajectory frame as a new current to-be-matched trajectory frame, and returning to execute step S210 until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed, obtaining a current alignment result corresponding to the current matching process, and continuing to execute step S270.

[0174] S270, sequentially traversing the temporarily-stored target trajectory frames that fail to match, for any one target trajectory frame that fails to match, obtaining the optimized pose of the target trajectory frame, and returning to execute step S210 until all temporarily-stored target trajectory frames that fail to match are traversed, obtaining a new current alignment result, and continuing to execute step S280.

[0175] S280, taking the driving trajectory corresponding to the new current alignment result as a new current reference trajectory, selecting another un-matched driving trajectory from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and returning to execute the matching operation of the new current reference trajectory and the new current to-be-matched trajectory, that is, returning to execute step S210 until all parking lot driving trajectories are traversed, obtaining a target matching result of the vehicle pose.

[0176] In the embodiment, after the matching of the current reference trajectory and the current to-be-matched trajectory is completed, many to-be-matched frames have achieved correct matching, and the relative pose between the to-be-matched trajectory and the reference trajectory is more accurate than before the matching. Based on this, by querying and matching the reference trajectory frame at the corresponding position again for the target trajectory frame that fails to match, the alignment result of the vehicle pose can be made more accurate. In addition, by taking the matching result obtained each time as a new reference trajectory and repeatedly executing the matching of the new reference trajectory and the new to-be-matched trajectory, accurate association of the multiple collected driving trajectories is achieved. Compared with the way of manually associating the multiple collected parking lot data in the related art, the method provided in the embodiment further improves the efficiency and accuracy of the association of the parking lot data.

[0177] Embodiment Three

[0178] Figure 3 A structural block diagram of a parking lot data matching device provided for the third embodiment of the application is shown in FIG. 3, which includes a driving trajectory selection module 310, a matching module 320, and a data association module 330, wherein, Figure 3

[0179] ​The driving track selection module 310 is configured to randomly select different vehicle driving tracks from the multiple collected parking lot driving tracks, and take one of the driving tracks as a current reference track and another driving track as a current to-be-matched track, wherein each driving track comprises a pre-parking lot track, an in-parking lot track and a post-parking lot track.

[0180] The matching module 320 is configured to match the current reference track with the current to-be-matched track. In the matching process, based on the alignment frame in which the vehicle poses in the current reference track and the current to-be-matched track are matched, the relative poses between each to-be-matched track frame in the current to-be-matched track, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, the poses of each to-be-matched track frame are adjusted in turn, and the pose-adjusted track frame is aligned with the corresponding reference track frame in the current reference track to obtain a current alignment result corresponding to the current matching process.

[0181] The data association module 330 is configured to take the driving track corresponding to the current alignment result as a new current reference track, select another un-matched driving track from the multiple collected parking lot driving tracks as a new current to-be-matched track, and return to perform a matching operation of the new current reference track and the new current to-be-matched track until a predetermined number of parking lot driving track matching operations are completed to obtain a target matching result of the vehicle pose.

[0182] Optionally, the matching module 320 comprises:

[0183] The current reference track frame determination unit is configured to traverse each to-be-matched track frame in the current to-be-matched track in turn, and for each current to-be-matched track frame, determine a current reference track frame corresponding to the position of the current to-be-matched track frame in the current reference track.

[0184] The matching unit is configured to match the current reference track frame with the current to-be-matched track frame.

[0185] The relative pose determination unit is configured to, if the current reference track frame and the current to-be-matched track frame are successfully matched, determine a first relative pose between the current to-be-matched track frame and the corresponding current reference track frame.

[0186] The first pose optimization unit is configured to, based on the first relative pose, the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot in each driving track, and the second relative poses between each to-be-matched track frame in the current to-be-matched driving track, perform pose optimization on all successfully matched to-be-matched track frames, and take the pose-optimized to-be-matched track frame and the reference track frame corresponding to the matched position as an alignment frame.

[0187] The pose alignment unit is configured to adjust the pose of the to-be-matched trajectory frame adjacent to the current alignment frame based on the current alignment frame, take the pose-adjusted adjacent trajectory frame as a new current to-be-matched trajectory frame, and return to perform the matching operation of the new current to-be-matched trajectory frame and the current reference trajectory frame corresponding to the position until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed, and a current alignment result of the vehicle pose in the current matching process is obtained.

[0188] Optionally, the device provided by the embodiment of the application further comprises:

[0189] The second pose optimization unit is configured to, after determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, perform pose optimization on all reference trajectory frames in the current reference trajectory based on the first relative pose, absolute position information of the vehicle in each driving trajectory before entering and after leaving the parking lot, and second relative poses between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, and take the pose-optimized to-be-matched trajectory frame and the pose-optimized reference trajectory frame matched successfully as alignment frames.

[0190] Optionally, the matching unit comprises:

[0191] The local feature information acquisition subunit is configured to acquire local feature information of the parking lot, and the local feature information comprises traffic sign, wall identification, road surface identification, and lane line identification.

[0192] The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame; or

[0193] The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame; or

[0194] Optionally, the device provided by the embodiment of the application further comprises:

[0195] The matching failure frame temporary storage module is configured to, if the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully, temporarily store the current to-be-matched trajectory frame matched unsuccessfully.

[0196] The re-matching module is configured to sequentially traverse the temporarily-stored target trajectory frames with failed matching, and repeatedly perform the matching operation of the target trajectory frame with the optimized pose and the reference trajectory frame corresponding to the position based on the optimized pose of the target trajectory frame with failed matching until all the temporarily-stored target trajectory frames with failed matching are traversed, so as to obtain a new current alignment result;

[0197] Correspondingly, the data association module is specifically configured to:

[0198] take the driving trajectory corresponding to the new current alignment result as a new current reference trajectory, select another un-matched driving trajectory from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and return to perform the matching operation of the new current reference trajectory and the new current to-be-matched trajectory until a predetermined number of parking lot driving trajectory matching operations are completed, so as to obtain a target matching result of the vehicle pose.

[0199] Optionally, the current reference trajectory frame determination unit comprises:

[0200] The first search subunit is configured to, in the case that the current to-be-matched trajectory frame belongs to the pre-parking lot trajectory or the post-parking lot trajectory, acquire the absolute position corresponding to the current to-be-matched trajectory frame, and search in the current reference trajectory within a first set distance range from the absolute position to obtain the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame.

[0201] Alternatively,

[0202] The second search subunit is configured to, in the case that the current to-be-matched trajectory frame belongs to the in-parking lot trajectory, search in the current reference trajectory within a first set distance range from the current position information corresponding to the current to-be-matched trajectory frame to obtain a candidate reference trajectory frame.

[0203] The same layer judgment subunit is configured to judge whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same layer of the parking lot.

[0204] The current reference trajectory frame determination subunit is configured to, if the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same layer of the parking lot, take the candidate reference trajectory frame as the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame.

[0205] Optionally, the same layer judgment subunit comprises:

[0206] The shortest topological path determination component is configured to determine the shortest topological path from the current to-be-matched trajectory frame to the candidate reference trajectory frame.

[0207] a slope road section detection component configured to detect whether there is a slope road section in the shortest topological path;

[0208] a same floor determination component configured to determine that the candidate reference trajectory frame and the current trajectory frame to be matched belong to a same floor of the parking lot if there is no slope road section in the shortest topological path; or,

[0209] determine that the candidate reference trajectory frame and the current trajectory frame to be matched belong to a same floor of the parking lot if there is a slope road section in the shortest topological path and if a difference between the height information of the uphill road section and the height information of the downhill road section in the slope road section is within a second set distance range; or,

[0210] determine that the candidate reference trajectory frame and the current trajectory frame to be matched do not belong to a same floor of the parking lot if the difference between the height information of the uphill road section and the height information of the downhill road section in the slope road section exceeds the second set distance range.

[0211] Optionally, the same floor determination component is specifically configured to:

[0212] determine whether a pitch angle of the vehicle relative to a horizontal plane is greater than a preset angle;

[0213] determine a road section corresponding to a trajectory frame with a pitch angle greater than or equal to the preset angle as a slope road section, and determine a road section corresponding to a trajectory frame with a pitch angle less than the preset angle as a non-slope road section.

[0214] Optionally, the first pose optimization unit is specifically configured to:

[0215] construct a first topological graph corresponding to the current trajectory to be matched and a second topological graph corresponding to the current reference trajectory, wherein nodes of the first topological graph are the trajectory frames to be matched, adjacent trajectory frames to be matched are connected by edges, nodes of the second topological graph are the reference trajectory frames, and adjacent reference trajectory frames are connected by edges;

[0216] optimize the first topological graph by a pose graph optimization method based on the first relative pose, absolute position information of the vehicle before entering the parking lot and after leaving the parking lot in each driving trajectory, and second relative poses between the trajectory frames to be matched in the current trajectory to be matched, and use the trajectory frames to be matched after pose optimization and the reference trajectory frames matched successfully as alignment frames, wherein the alignment frames in the first topological graph and the second topological graph are connected by edges.

[0217] Optionally, the apparatus provided in the embodiment further includes:

[0218] The initial optimization module is configured to optimize the first topological graph and the second topological graph respectively based on the absolute position information before entering the parking lot and after leaving the parking lot, so as to reduce the deviation between the current to-be-matched trajectory and the current reference trajectory in the absolute position.

[0219] Optionally, the vehicle pose corresponding to each reference trajectory frame or the vehicle pose corresponding to each to-be-matched trajectory frame is obtained by any one of the following sensor data, or is obtained by fusing a plurality of the following sensor data.

[0220] Inertial measurement unit (IMU) data, image data, radar point cloud data, and odometry data.

[0221] The parking lot data matching device provided in the embodiments of the present application can perform the parking lot data matching method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method. Technical details not described in detail in the above embodiments can be referred to the parking lot data matching method provided in any of the embodiments of the present application.

[0222] Embodiment Four

[0223] Figure 4 A structural block diagram of an electronic device provided in the fourth embodiment of the present application is shown in FIG. 4, which includes: Figure 4

[0224] a memory 510 storing executable program codes;

[0225] a processor 520 coupled with the memory 510;

[0226] The processor 520 calls the executable program codes stored in the memory 510 to execute the parking lot data matching method provided in any of the embodiments of the present application.

[0227] Embodiment Five

[0228] Based on the above embodiments, another embodiment of the present application provides a vehicle, which includes the device as described in any of the above embodiments, or includes the electronic device as described above.

[0229] Figure 5 A schematic diagram of a vehicle provided in the fifth embodiment of the present application is shown in FIG. 5, which includes: Figure 5 ​As shown, the vehicle includes a speed sensor 61, an ECU (Electronic Control Unit) 62, a GPS (Global Positioning System) positioning device 63, and a T-Box (Telematics Box) 64. Among them, the speed sensor 61 is used to measure the vehicle speed and provide the experience speed for model training; the GPS positioning device 63 is used to obtain the current geographic position of the vehicle; the T-Box 64 can communicate with the server as a gateway; and the ECU 62 can execute the matching method of the parking lot data.

[0230] In addition, the vehicle can further include a V2X (Vehicle-to-Everything) module 65, a radar 66, and a camera 67. The V2X module 65 is used to communicate with other vehicles, roadside devices, and the like; the radar 66 or the camera 67 is used to perceive the road environment information in front and / or other directions to obtain raw point cloud data; and the radar 66 and / or the camera 67 can be configured at the front of the vehicle body and / or the tail of the vehicle body.

[0231] Based on the above method embodiments, another embodiment of the present application provides a storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to implement the matching method of the parking lot data according to any one of the above embodiments.

[0232] Those skilled in the art can understand that the drawings are only schematic diagrams of an embodiment, and the modules or flows in the drawings are not necessarily required for implementing the present application.

[0233] Those skilled in the art can understand that the modules in the device in the embodiments can be distributed in the device in the embodiments according to the embodiment description, or can be changed and located in one or more devices different from the embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0234] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A parking lot data matching method characterized by, The method comprises the following steps: arbitrarily selecting different vehicle driving trajectories from multiple collected parking lot driving trajectories, taking one of the driving trajectories as a current reference trajectory and taking another driving trajectory as a current to-be-matched trajectory, wherein each driving trajectory comprises a pre-parking-lot trajectory, an in-parking-lot trajectory and a post-parking-lot trajectory; matching the current reference trajectory with the current to-be-matched trajectory, wherein in the matching process, based on the aligned frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory match each other, the relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, the poses of each to-be-matched trajectory frame are adjusted in sequence, and the trajectory frame after the pose adjustment is aligned with the corresponding reference trajectory frame in the current reference trajectory to obtain a current alignment result corresponding to the current matching process; taking the driving trajectory corresponding to the current alignment result as a new current reference trajectory, selecting another un-matched driving trajectory from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and returning to perform the matching operation of the new current reference trajectory and the new current to-be-matched trajectory until a predetermined number of parking lot driving trajectory matching operations are completed to obtain a target matching result of the vehicle pose.

2. The method of claim 1, wherein, The matching of the current reference trajectory with the current to-be-matched trajectory, wherein in the matching process, based on the aligned frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory match each other, the relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, the poses of each to-be-matched trajectory frame are adjusted in sequence, and the trajectory frame after the pose adjustment is aligned with the corresponding reference trajectory frame in the current reference trajectory to obtain a current alignment result corresponding to the current matching process, comprises the following steps: sequentially traversing each to-be-matched trajectory frame in the current to-be-matched trajectory, for each current to-be-matched trajectory frame, determining a current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame in the current reference trajectory; matching the current reference trajectory frame with the current to-be-matched trajectory frame; if the matching is successful, determining a first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame; based on the first relative pose, the absolute position information of the vehicle before entering the parking lot and after leaving the parking lot in each driving trajectory, and a second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, performing pose optimization on all to-be-matched trajectory frames that have been successfully matched, and taking the to-be-matched trajectory frame after the pose optimization and the reference trajectory frame corresponding to the matching position as an alignment frame; based on the current alignment frame, adjusting the poses of the to-be-matched trajectory frames adjacent to the current alignment frame, taking the to-be-matched trajectory frames after the pose adjustment as new current to-be-matched trajectory frames, and returning to perform the matching operation of the new current to-be-matched trajectory frames with the current reference trajectory frames corresponding to the positions until all to-be-matched trajectory frames in the current to-be-matched trajectory are traversed to obtain a current alignment result of the vehicle pose in the current matching process.

3. The method of claim 2, wherein, After determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, the method further comprises: performing pose optimization on all reference trajectory frames in the current reference trajectory based on the first relative pose, absolute position information of the vehicle in each driving trajectory before entering and after exiting the parking lot, and second relative poses between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, and taking the to-be-matched trajectory frame after pose optimization and the successfully matched reference trajectory frame after pose optimization as an aligned frame.

4. The method of claim 2, wherein, The matching of the current reference trajectory frame and the current to-be-matched trajectory frame comprises: acquiring local feature information of the parking lot, wherein the local feature information comprises traffic signs, wall identification, road surface identification, and lane line identification; if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are successfully matched; or if the local feature information of the parking lot corresponding to the current reference trajectory frame is inconsistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame, it is determined that the current reference trajectory frame and the current to-be-matched trajectory frame are unsuccessfully matched.

5. The method according to any of claims 2-4, characterized by, The method further comprises: if the current reference trajectory frame and the current to-be-matched trajectory frame are unsuccessfully matched, temporarily storing the unsuccessfully matched current to-be-matched trajectory frame; after obtaining the current alignment result, sequentially traversing the temporarily stored unsuccessfully matched target trajectory frames, for any one of the unsuccessfully matched target trajectory frames, repeatedly performing the matching operation of the target trajectory frame after pose optimization and the reference trajectory frame corresponding in position based on the pose of the target trajectory frame after optimization, until all the temporarily stored unsuccessfully matched target trajectory frames are traversed, and a new current alignment result is obtained; correspondingly, the driving trajectory corresponding to the current alignment result is taken as a new current reference trajectory, another un-matched driving trajectory is selected from the multiple collected driving trajectories of the parking lot as a new current to-be-matched trajectory, and the matching operation of the new current reference trajectory and the new current to-be-matched trajectory is performed again, comprising: the driving trajectory corresponding to the new current alignment result is taken as a new current reference trajectory, another un-matched driving trajectory is selected from the multiple collected driving trajectories of the parking lot as a new current to-be-matched trajectory, and the matching operation of the new current reference trajectory and the new current to-be-matched trajectory is performed again.

6. The method of claim 2, wherein, The determination of the current reference trajectory frame corresponding in position to the current to-be-matched trajectory frame in the current reference trajectory comprises: in the case that the current to-be-matched trajectory frame belongs to a pre-entering-parking-lot trajectory or a post-exiting-parking-lot trajectory: an absolute position corresponding to the current to-be-matched trajectory frame is acquired, and a search is performed within a first set distance range from the absolute position in the current reference trajectory to obtain a current reference trajectory frame corresponding in position to the current to-be-matched trajectory frame; or, in the case that the current to-be-matched trajectory frame belongs to an in-parking-lot trajectory: searching in the current reference trajectory within a first set distance range from the current position information corresponding to the current to-be-matched trajectory frame, to obtain a candidate reference trajectory frame; determining whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot; if they belong to the same floor, taking the candidate reference trajectory frame as the current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame.

7. The method of claim 6, wherein, The determination of whether the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot includes: determining a shortest topological path from the current to-be-matched trajectory frame to the candidate reference trajectory frame; detecting whether there is a slope road section in the shortest topological path; if there is no slope road section, determining that the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot; or if there is a slope road section, and if the difference between the height information of the uphill section and the height information of the downhill section in the slope road section is within a second set distance range, determining that the candidate reference trajectory frame and the current to-be-matched trajectory frame belong to the same floor of the parking lot; if the difference between the height information of the uphill section and the height information of the downhill section in the slope road section exceeds the second set distance range, determining that the candidate reference trajectory frame and the current to-be-matched trajectory frame do not belong to the same floor of the parking lot.

8. The method of claim 7, wherein, The detection of whether there is a slope road section in the shortest topological path includes: determining whether the pitch angle of the vehicle relative to the horizontal plane is greater than a preset angle; taking a road section corresponding to a trajectory frame with a pitch angle greater than or equal to the preset angle as a slope road section, and taking a road section corresponding to a trajectory frame with a pitch angle less than the preset angle as a non-slope road section.

9. The method of claim 2, wherein, The pose optimization of all to-be-matched trajectory frames that have been successfully matched based on the first relative pose, the absolute position information of the vehicle before entering and after leaving the parking lot in each driving trajectory, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory includes: constructing a first topological graph corresponding to the current to-be-matched trajectory and a second topological graph corresponding to the current reference trajectory, wherein the nodes of the first topological graph are each to-be-matched trajectory frame, and adjacent to-be-matched trajectory frames are connected by edges, and the nodes of the second topological graph are each reference trajectory frame, and adjacent reference trajectory frames are connected by edges; optimizing the first topological graph by a pose graph optimization method based on the first relative pose, the absolute position information of the vehicle before entering and after leaving the parking lot in each driving trajectory, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving trajectory, and taking the to-be-matched trajectory frame after pose optimization and the successfully matched reference trajectory frame as an alignment frame, wherein the nodes of the alignment frame in the first topological graph and the second topological graph are connected by edges.

10. The method of claim 9, wherein, Before matching the current reference trajectory and the current to-be-matched trajectory, the method further includes: The first topological graph and the second topological graph are respectively optimized based on absolute position information before entering the parking lot and after leaving the parking lot, so as to reduce deviation between the current to-be-matched trajectory and the current reference trajectory in absolute position.

11. The method of claim 1, wherein, The vehicle pose corresponding to each reference trajectory frame or the vehicle pose corresponding to each to-be-matched trajectory frame is obtained by any one of the following sensor data or by fusing multiple types of sensor data: Inertial measurement unit (IMU) data, image data, radar point cloud data, and odometry data.

12. A parking lot data matching device characterized by comprising: Comprise: a driving trajectory selection module configured to select any one of the vehicle driving trajectories collected at different times from the multiple collected parking lot driving trajectories, and take one of the driving trajectories as a current reference trajectory and take another driving trajectory as a current to-be-matched trajectory, wherein each driving trajectory comprises a trajectory before entering the parking lot, a trajectory in the parking lot, and a trajectory after leaving the parking lot; a matching module configured to match the current reference trajectory with the current to-be-matched trajectory, wherein in the matching process, based on aligned frames in which the vehicle poses in the current reference trajectory and the current to-be-matched trajectory are matched, relative poses between each to-be-matched trajectory frame in the current to-be-matched trajectory, and absolute position information of the vehicle before entering the parking lot and after leaving the parking lot, the pose of each to-be-matched trajectory frame is adjusted in turn, and the trajectory frame after the pose adjustment is aligned with the corresponding reference trajectory frame in the current reference trajectory to obtain a current alignment result corresponding to the current matching process; a data association module configured to take the driving trajectory corresponding to the current alignment result as a new current reference trajectory, select another unmatched driving trajectory from the multiple collected parking lot driving trajectories as a new current to-be-matched trajectory, and return to perform a matching operation of the new current reference trajectory and the new current to-be-matched trajectory until a predetermined number of parking lot driving trajectory matching operations are completed to obtain a target matching result of the vehicle pose.

13. The apparatus of claim 12, wherein, The matching module comprises: a current reference trajectory frame determination unit configured to match the current reference trajectory with the current to-be-matched trajectory, wherein in the matching process, each to-be-matched trajectory frame in the current to-be-matched trajectory is sequentially traversed, and for each current to-be-matched trajectory frame, a current reference trajectory frame corresponding to the position of the current to-be-matched trajectory frame is determined in the current reference trajectory; a matching unit configured to match the current reference trajectory frame with the current to-be-matched trajectory frame; a relative pose determination unit configured to determine a first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame if the current reference trajectory frame and the current to-be-matched trajectory frame are successfully matched. The first pose optimization unit is configured to perform pose optimization on all matched trajectory frames based on the first relative pose, absolute position information of the vehicle before entering and after leaving the parking lot in each driving track, and the second relative pose between each trajectory frame in the current to-be-matched driving track, and take the to-be-matched trajectory frame after pose optimization and the reference trajectory frame corresponding to the matched position as an alignment frame. The pose alignment unit is configured to adjust the pose of a to-be-matched trajectory frame adjacent to the current alignment frame based on the current alignment frame, take the adjacent trajectory frame after pose adjustment as a new current to-be-matched trajectory frame, and return to perform a matching operation of the new current to-be-matched trajectory frame and a current reference trajectory frame corresponding to a position until all to-be-matched trajectory frames in the current to-be-matched track are traversed, and a current alignment result of the vehicle pose in the current matching process is obtained.

14. The apparatus of claim 13, wherein, The device further comprises: The second pose optimization unit is configured to perform pose optimization on all reference trajectory frames in the current reference track based on the first relative pose, absolute position information of the vehicle before entering and after leaving the parking lot in each driving track, and the second relative pose between each to-be-matched trajectory frame in the current to-be-matched driving track after determining the first relative pose between the current to-be-matched trajectory frame and the corresponding current reference trajectory frame, and take the to-be-matched trajectory frame after pose optimization and the reference trajectory frame after pose optimization matched successfully as an alignment frame.

15. The apparatus of claim 13, wherein, The matching unit comprises: The local feature information acquisition subunit is configured to acquire local feature information of the parking lot, and the local feature information comprises traffic signs, wall identification, road surface identification, and lane line identification. The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame; or The matching subunit is configured to determine that the current reference trajectory frame and the current to-be-matched trajectory frame are matched successfully if the local feature information of the parking lot corresponding to the current reference trajectory frame is consistent with the local feature information of the parking lot corresponding to the current to-be-matched trajectory frame; or 16. The apparatus of any of claims 13-15, wherein, The device further comprises: The matching failure frame temporary storage module is configured to temporarily store the current to-be-matched trajectory frame matched unsuccessfully if the current reference trajectory frame and the current to-be-matched trajectory frame are matched unsuccessfully. The re-matching module is configured to traverse the temporarily stored target trajectory frame matched unsuccessfully in sequence, repeatedly perform a matching operation of the target trajectory frame after pose optimization and a reference trajectory frame corresponding to a position based on the pose of the target trajectory frame after optimization until all temporarily stored target trajectory frames matched unsuccessfully are traversed, and a new current alignment result is obtained. Correspondingly, the data association module is specifically configured to: The new current alignment result corresponds to a driving track as a new current reference track, and another unmatched driving track is selected from the multiple collected parking lot driving tracks as a new current to-be-matched track, and the matching operation of the new current reference track and the new current to-be-matched track is returned to be performed until a predetermined number of parking lot driving track matching operations are completed, and a target matching result of the vehicle pose is obtained.

17. The apparatus of claim 13, wherein, The current reference track frame determination unit comprises: The first search subunit is configured to, in the case that the current to-be-matched track frame belongs to the pre-parking lot track or the post-parking lot track, acquire an absolute position corresponding to the current to-be-matched track frame, and search in the current reference track within a first set distance range from the absolute position to obtain a current reference track frame corresponding to the position of the current to-be-matched track frame; Or, The second search subunit is configured to, in the case that the current to-be-matched track frame belongs to the in-parking lot track, search in the current reference track within a first set distance range from the current position information corresponding to the current to-be-matched track frame to obtain a candidate reference track frame; The same layer judgment subunit is configured to judge whether the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot; The current reference track frame determination subunit is configured to, if the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot, take the candidate reference track frame as the current reference track frame corresponding to the position of the current to-be-matched track frame.

18. The apparatus of claim 17, wherein, The same layer judgment subunit comprises: The shortest topological path determination component is configured to determine a shortest topological path from the current to-be-matched track frame to the candidate reference track frame; The slope road section detection component is configured to detect whether there is a slope road section in the shortest topological path; The same layer determination component is configured to, if there is no slope road section in the shortest topological path, determine that the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot; Or, if there is a slope road section in the shortest topological path, and if the difference between the height information of the uphill road section and the height information of the downhill road section in the slope road section is within a second set distance range, it is determined that the candidate reference track frame and the current to-be-matched track frame belong to the same layer of the parking lot; Or, if the difference between the height information of the uphill road section and the height information of the downhill road section in the slope road section exceeds the second set distance range, it is determined that the candidate reference track frame and the current to-be-matched track frame do not belong to the same layer of the parking lot.

19. The apparatus of claim 18, wherein, The same layer determination component is specifically configured to: Judge whether the pitch angle of the vehicle relative to the horizontal plane is greater than a preset angle; The road section corresponding to the track frame with the pitch angle greater than or equal to the preset angle is taken as a slope road section, and the road section corresponding to the track frame with the pitch angle less than the preset angle is taken as a non-slope road section.

20. The apparatus of claim 13, wherein, The first pose optimization unit is specifically configured to: constructing a first topology graph corresponding to the current to-be-matched trajectories and a second topology graph corresponding to the current reference trajectories, wherein the nodes of the first topology graph are the to-be-matched trajectory frames, and the adjacent to-be-matched trajectory frames are connected by edges, and the nodes of the second topology graph are the reference trajectory frames, and the adjacent reference trajectory frames are connected by edges; based on the first relative poses, the absolute position information of the vehicle in each driving trajectory before entering and after leaving the parking lot, and the second relative poses between the to-be-matched trajectory frames in the current to-be-matched driving trajectory, the first topology graph is optimized by a pose graph optimization method, and the to-be-matched trajectory frames after pose optimization and the matched reference trajectory frames are used as alignment frames, wherein the nodes of the alignment frames in the first topology graph and the second topology graph are connected by edges.

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

22. A storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the method of any one of claims 1-11.

23. A vehicle characterized by comprising: The vehicle comprises the method of any one of claims 1-11, or comprises the electronic device of claim 21.

Images