Parking track matching method and device

Abstract

The invention discloses a parking track matching method and device. The method comprises the following steps: acquiring driving positions and environment characteristics corresponding to driving track points of N groups of vehicles; carrying out coding storage processing on the N groups of driving positions and the environment features to obtain a driving track data set corresponding to the driving track segment of the vehicle; according to the driving track data set, performing rough matching on the driving track segment and mapping track points in a parking map pre-stored in the vehicle, and determining k candidate track segments from the mapping track points of the parking map; and performing fine matching on the driving track section and the k candidate track sections, and determining a target track section from the k candidate track sections to determine the position of the vehicle in the parking map. Thus, by means of the rough matching strategy of the track point environment characteristics, the mismatching occurrence probability is reduced, the reliability of the track matching result is improved, finally, by means of the fine matching strategy, the track matching precision is improved, and the parking track matching success rate of the vehicle in the basement scene is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a parking trajectory matching method and device. Background Art

[0002] As the popularity of smart cars continues to accelerate, users' requirements for diversified functions of smart cars are increasing dramatically, especially the parking function in parking lots, which is one of the essential functions of smart cars nowadays.

[0003] The success rate of the underground parking function has a direct impact on the reputation of smart cars and the intelligence level of the OEM. The main reason affecting the launch of underground parking functions is the repositioning of parking. Since there is no global information in the underground parking lot and the positioning signal is weak, the vehicle's pre-position is inaccurate, which leads to the failure of repositioning and makes the parking function unusable.

[0004] Therefore, how to efficiently and accurately match parking trajectories is a technical problem that we urgently need to solve. Summary of the invention

[0005] In view of the above problems, the present invention provides a parking trajectory matching method and device, which effectively solves the technical problem of low efficiency and accuracy of trajectory matching in garage parking relocation.

[0006] According to a first aspect of the present invention, a parking trajectory matching method is provided, comprising:

[0007] Obtain the driving positions and environmental characteristics corresponding to the driving trajectory points of N groups of vehicles;

[0008] Encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle;

[0009] According to the driving trajectory data set corresponding to the driving trajectory segment, the driving trajectory segment is roughly matched with the mapping trajectory points in the parking map pre-stored in the vehicle, and k candidate trajectory segments are determined from the mapping trajectory points in the parking map, each candidate trajectory segment is composed of N continuous mapping trajectory points;

[0010] The driving trajectory segment is precisely matched with the k candidate trajectory segments, and a target trajectory segment is determined from the k candidate trajectory segments to determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

[0011] Optionally, encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle includes:

[0012] For each driving trajectory point, attribute encoding is performed on the environmental feature corresponding to the driving trajectory point to obtain the environmental attribute information corresponding to the driving trajectory point;

[0013] The driving position and the environmental attribute information corresponding to the driving trajectory point are used as four-dimensional driving trajectory data, wherein N groups of the driving trajectory data constitute the driving trajectory data set;

[0014] Performing validity detection on the driving trajectory data in the driving trajectory data set;

[0015] If the number of valid driving trajectory data in the driving trajectory data set is greater than the number threshold, a rough matching process is performed.

[0016] Optionally, the roughly matching the driving trajectory segment with mapping trajectory points in a parking map pre-stored in the vehicle according to the driving trajectory data set corresponding to the driving trajectory segment, and determining k candidate trajectory segments from the mapping trajectory points in the parking map, comprises:

[0017] Obtain the vehicle's driving posture information;

[0018] Putting the mapping trajectory data corresponding to the mapping trajectory points in the parking map into KDTree, and searching for the mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle;

[0019] If a mapping trajectory point in the parking map is searched within the target radius, a matching trajectory point set is determined according to the sequence number of the searched mapping trajectory point;

[0020] In the matching trajectory point set, starting from the matching trajectory point with the smallest serial number, the matching trajectory point is used as the starting trajectory point, and extraction processing is performed to obtain m matching trajectory segments, and the extraction processing step includes: starting from the starting trajectory point, sequentially extracting N matching trajectory points as a matching trajectory segment; m is an integer greater than 1;

[0021] Traversing all matching trajectory segments, and respectively calculating the environmental attribute error between each matching trajectory segment and the driving trajectory segment;

[0022] According to the environmental attribute errors of the matching trajectory segments, k candidate trajectory segments are determined from the matching trajectory segments.

[0023] Optionally, after the step of searching for mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle, the method further includes:

[0024] If no mapping trajectory point in the parking map is found within the target radius, the target radius is expanded and updated to obtain an updated target radius, and the step of searching for the mapping trajectory point in the parking map within the target radius is returned to be executed; or,

[0025] If no mapping trajectory point in the parking map is found within the target radius, the process returns to the step of obtaining the driving positions and environmental features corresponding to the driving trajectory points of the N groups of vehicles.

[0026] Optionally, the precisely matching the driving trajectory segment with the k candidate trajectory segments and determining a target trajectory segment from the k candidate trajectory segments to determine the position of the vehicle in the parking map includes:

[0027] For each candidate trajectory segment, respectively calculating a total error between the candidate trajectory segment and the driving trajectory segment;

[0028] Determine a target trajectory segment matching the driving trajectory segment from the k candidate trajectory segments according to a total error between each candidate trajectory segment and the driving trajectory segment;

[0029] Get the recursive posture information of the vehicle at the current moment;

[0030] Determine the current target posture information of the vehicle according to the posture information corresponding to the mapping trajectory point with the largest sequence number in the target trajectory segment, the recursive posture information and the driving posture information of the vehicle; wherein the driving posture information is the posture information obtained when the vehicle starts to perform rough matching, the recursive posture information is the posture information at the current moment derived based on the driving posture information, and the target posture information is the posture information calculated after the vehicle performs fine matching processing;

[0031] The position of the vehicle in the parking map is determined according to the target posture information.

[0032] Optionally, for each candidate trajectory segment, respectively calculating a total error between the candidate trajectory segment and the driving trajectory segment includes:

[0033] Calculating the driving center point of the driving trajectory segment;

[0034] For each driving trajectory point in the driving trajectory segment, according to the driving center point, calculate a target driving trajectory point corresponding to the driving trajectory point;

[0035] Calculate the candidate center point of the candidate trajectory segment;

[0036] For each mapping trajectory point in the candidate trajectory segment, calculate the target mapping trajectory point corresponding to the mapping trajectory point according to the candidate center point;

[0037] According to the target driving trajectory point and the target mapping trajectory point, a residual function of the candidate trajectory segment and the driving trajectory segment with respect to rotation and translation is constructed, and a transformation matrix from the driving trajectory segment to the candidate trajectory segment is calculated;

[0038] According to the transformation matrix, the driving trajectory segment is rotated and transformed to obtain a rotation trajectory segment of the driving trajectory segment with respect to the candidate trajectory segment;

[0039] The total error of the driving trajectory segment with respect to the candidate trajectory segment is calculated according to the rotation trajectory segment and the candidate trajectory segment.

[0040] Optionally, the obtaining of the positions of multiple groups of vehicles and environmental features corresponding to the trajectory points of the vehicles includes:

[0041] Get the vehicle's location;

[0042] If the position of the vehicle is within a preset distance of the area where the parking garage is located, multiple groups of vehicle positions and environmental features corresponding to the trajectory points of the vehicle's travel are obtained; wherein a parking map corresponding to the parking garage is pre-stored in the vehicle.

[0043] According to a second aspect of the present invention, a parking trajectory matching device is provided, comprising:

[0044] An acquisition module, used to acquire the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles;

[0045] An encoding module, used for encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle, to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle;

[0046] a coarse matching module, configured to coarsely match the driving trajectory segment with the mapping trajectory points in the parking map pre-stored in the vehicle according to the driving trajectory data set corresponding to the driving trajectory segment, and determine k candidate trajectory segments from the mapping trajectory points in the parking map, each candidate trajectory segment consisting of N consecutive mapping trajectory points;

[0047] A precise matching module is used to precisely match the driving trajectory segment with the k candidate trajectory segments, determine a target trajectory segment from the k candidate trajectory segments, and determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

[0048] According to a third aspect of the present invention, a controller is provided. The controller includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the aforementioned parking trajectory matching method.

[0049] According to a fourth aspect of the present invention, a vehicle is provided, the vehicle comprising a vehicle body and a controller installed in the vehicle body, wherein the controller executes the aforementioned parking trajectory matching method.

[0050] The above one or more technical solutions in the embodiments of this specification have at least the following technical effects:

[0051] The embodiment of the present specification provides a parking trajectory matching method and device, which obtains the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles; encodes and stores the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles to obtain the driving trajectory data set corresponding to the driving trajectory segment of the vehicle; according to the driving trajectory data set corresponding to the driving trajectory segment, the driving trajectory segment is roughly matched with the mapping trajectory points in the parking map pre-stored in the vehicle, and k candidate trajectory segments are determined from the mapping trajectory points of the parking map, each candidate trajectory segment is composed of N continuous mapping trajectory points; the driving trajectory segment is precisely matched with the k candidate trajectory segments, and the target trajectory segment is determined from the k candidate trajectory segments to determine the position of the vehicle in the parking map. In this way, the rough matching strategy of the trajectory point attributes is used to reduce the probability of mismatching, greatly improve the reliability of the trajectory matching results, and finally the precise matching strategy is used to not only improve the accuracy of trajectory matching, but also improve the matching effect, ensure the success rate of parking trajectory matching of vehicles in underground parking scenarios, and improve user experience.

[0052] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Also, throughout the accompanying drawings, the same reference figures are used to represent the same components. In the drawings:

[0054] Figure 1 A flow chart of a parking trajectory matching method in an embodiment of the present invention is shown.

[0055] Figure 2 A schematic diagram showing environmental attribute information in an embodiment of the present invention.

[0056] Figure 3 A block diagram of a parking trajectory matching device in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0058] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0060] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] When a driver drives a vehicle into a parking garage, for the first time entering the parking garage and may park in the garage many times in the future, the driver will generally create a parking map of the parking garage. When driving the vehicle to the parking garage again in the future, based on the pre-stored parking map, trajectory matching can be performed, and then the smart parking function can be launched. However, in actual situations, some parking garages are large, and the constructed parking map cannot cover all parking garages, that is, the parking map only constructs part of the map data of the parking garage. In addition, parking garages often have multiple entrances and exits. When entering the parking garage, the driver may not drive according to the initial mapping trajectory points in the parking map. And if the parking garage is located underground, the positioning signal is weak, which affects the efficiency and progress of the garage parking trajectory matching, and thus affects the online usage rate of the smart parking function.

[0062] Based on the above situation, the embodiment of the present invention provides a parking trajectory matching method, combining Figure 1 As shown in the flowchart, the parking trajectory matching method includes steps 101 to 104:

[0063] Step 101: Obtain driving positions and environmental features corresponding to driving trajectory points of N groups of vehicles;

[0064] In this embodiment, the parking map is map data pre-built based on the parking garage and stored in the vehicle. The vehicle demarcates a geometric frame around the parking garage (hereinafter referred to as the garage), and the vehicle acquires the vehicle's position in real time. When the vehicle's position is detected to be within a preset distance of the geometric frame (for example, the preset distance is 20 meters or 25 meters, etc.), it indicates that the vehicle is about to arrive at the garage and may enter the garage for parking.

[0065] At this time, the vehicle starts to obtain the driving positions and environmental features corresponding to N groups of driving trajectory points. The driving position and environmental features corresponding to one driving trajectory point are regarded as a group. It should be noted that when collecting trajectory points, this embodiment collects one driving trajectory point for every a meters traveled (for example, a is 1), and collects the position and environmental features corresponding to the driving trajectory point.

[0066] Among them, the environmental features are set based on the environmental characteristics of the garage entrance. Commonly, the garage entrance is generally equipped with ramps, turns, pillars, speed bumps, etc., and the above features may be set in different locations due to individual differences in garages. Therefore, the data can be combined with the driving trajectory points for trajectory matching. As an example, the environmental features of this embodiment may include turning, uphill and downhill, gates, intersections, zebra crossings, speed bumps, pillar information, etc.

[0067] The vehicle's position refers to the position (x, y, z) obtained by track recursion based on the inertial navigation unit after the vehicle is powered on.

[0068] Step 102: Encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle;

[0069] In this embodiment, for each driving trajectory point, the environmental feature corresponding to the driving trajectory point is attribute-encoded to obtain the environmental attribute information corresponding to the driving trajectory point.

[0070] For example, combining Figure 2As shown, using a 16-bit binary data, from bit 10 to bit 0, it is judged in turn whether there are pillars on the left and right sides of the driving trajectory point, whether the driving trajectory point is on the speed bump, whether the driving trajectory point is passing through the gate, whether the driving trajectory point is on the zebra crossing, whether the driving trajectory point is in the intersection, the turning state of the driving trajectory point (straight, left or right turn), and the trajectory ramp information (flat ground, uphill or downhill). According to the judgment result, the value of the corresponding bit is modified (1 for yes and 0 for no), and the other 11-15 bits are reserved, and the added environmental features can be encoded later.

[0071] The above environmental attribute information is converted into double type data and combined with the vehicle position (x, y, z) into a four-dimensional data body, namely the driving trajectory data. The driving trajectory data is then stored in a queue of a fixed size N (N can be in the range of 20-30) as a driving trajectory data set.

[0072] In the driving trajectory data set, the validity of the stored driving trajectory data is checked, and the driving trajectory points other than flat ground and straight driving are regarded as valid driving trajectory data. If the number of valid driving trajectory data is greater than the number threshold, the rough matching step is entered. Because in the driving trajectory point data, if there are only straight driving and flat ground, the longitudinal position cannot be constrained, which easily leads to distortion of the matching result, so the program will not be able to enter the subsequent trajectory matching step.

[0073] If the number of detected valid driving trajectory data is less than or equal to the quantity threshold, the step of returning to the embodiment to execute the step of obtaining the driving position and environmental characteristics corresponding to the driving trajectory point of the vehicle.

[0074] Step 103: according to the driving trajectory data set corresponding to the driving trajectory segment, roughly matching the driving trajectory segment with the mapping trajectory points in the parking map pre-stored in the vehicle, and determining k candidate trajectory segments from the mapping trajectory points in the parking map, each candidate trajectory segment consisting of N consecutive mapping trajectory points;

[0075] In this embodiment, the multiple groups of driving trajectory points obtained constitute the current driving trajectory segment of the vehicle. In order to determine whether the vehicle is currently in the parking map, it is necessary to match the driving trajectory points with the mapping trajectory points in the parking map. In order to reduce the situation of no match, this embodiment will perform a rough match. If there are multiple candidate trajectory segments in the parking map, it means that the position of the vehicle may be in the parking map. If there is no candidate trajectory segment that meets the requirements in the parking map, it means that the position of the vehicle is far away from the area corresponding to the parking map, and there is no need to continue the precise matching of the trajectory, thereby reducing data processing. Because the trajectory matching process is time-consuming, but the main thread cannot be affected, it is necessary to open up an independent thread for trajectory matching.

[0076] Specifically, the rough matching step may include:

[0077] Obtain the vehicle's driving posture information;

[0078] Putting the mapping trajectory data corresponding to the mapping trajectory points in the parking map into KDTree, and searching for the mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle;

[0079] If a mapping trajectory point in the parking map is searched within the target radius, a matching trajectory point set is determined according to the sequence number of the searched mapping trajectory point;

[0080] In the matching trajectory point set, starting from the matching trajectory point with the smallest serial number, the matching trajectory point is used as the starting trajectory point, and extraction processing is performed to obtain m matching trajectory segments, and the extraction processing step includes: starting from the starting trajectory point, sequentially extracting N matching trajectory points as a matching trajectory segment; m is an integer greater than 1;

[0081] Traversing all matching trajectory segments, and respectively calculating the environmental attribute error between each matching trajectory segment and the driving trajectory segment;

[0082] According to the environmental attribute errors of the matching trajectory segments, k candidate trajectory segments are determined from the matching trajectory segments.

[0083] It should be noted that the vehicle's driving posture information refers to the vehicle's posture information obtained at the time of entering the rough matching. The posture information includes position, roll, pitch, and yaw angle. Among them, the position here refers to the position coordinates obtained based on RTK (Real time kinematic, real-time differential positioning) or GNSS (Global navigation Satellite System, Global navigation satellite system).

[0084] The parking map is pre-established map data about the garage, and the mapping trajectory points refer to the trajectory points of the vehicle when establishing the parking map. The mapping trajectory data includes the location of the mapping trajectory points and the environmental attribute information corresponding to the mapping trajectory points.

[0085] The mapping trajectory data corresponding to the mapping trajectory points in the parking map are placed in KDTree, and the driving posture information of the vehicle is used to search for the mapping trajectory points in the parking map within a target radius (for example, 25 meters). Since the driving posture information has errors, it is necessary to use KDTree to retrieve possible matching trajectory points.

[0086] It should be noted that if no mapping trajectory point in the parking map is found within the target radius, one way is to expand and update the target radius to obtain an updated target radius, and then return to and repeat the step of searching for the mapping trajectory point in the parking map within the target radius; generally, one or two expansion updates can be performed, the number of expansion updates should not be too many, and the updated target radius should not be too large. If no mapping trajectory point in the parking map is found within the target radius after the expansion update, the coarse matching operation is terminated, and the vehicle's position and environmental features need to be re-acquired and stored.

[0087] Another way is that if no mapping trajectory point in the parking map is found within the target radius, the target radius is not enlarged and updated, and the process directly returns to the step of obtaining the driving positions and environmental features corresponding to the driving trajectory points of the N groups of vehicles. That is, the vehicle is far away from the location of the parking map, and the process waits for the vehicle to travel a certain distance.

[0088] If the mapping trajectory points in the parking map are searched within the target radius, the matching trajectory point set is determined according to the maximum and minimum sequence numbers of the mapping trajectory points searched. Because the sequence numbers of all mapping trajectory points searched within the target radius are not necessarily continuous, but when the parking map is established, the mapping trajectory points are continuous. Therefore, this embodiment stores the mapping trajectory points with the smallest sequence number to the mapping trajectory points with the largest sequence number in the matching trajectory point set.

[0089] For example, after searching for mapping trajectory points with serial numbers 3, 4, 7, and 12, 11 consecutive mapping trajectory points with serial numbers 2-12 constitute a matching trajectory point set.

[0090] In the matching trajectory point set, starting from the mapping trajectory point with the smallest sequence number, N consecutive mapping trajectory points are extracted each time as a matching trajectory segment, and multiple matching trajectory segments can be obtained in the end.

[0091] When calculating the environmental attribute error between the matching trajectory segment and the driving trajectory segment, first convert the environmental attribute information corresponding to each mapping trajectory point of the matching trajectory segment into a decimal value, then convert the environmental attribute information corresponding to each driving trajectory point of the driving trajectory segment into a decimal value, and finally calculate the difference in environmental attribute information of each trajectory point, and then sum up the difference in environmental attribute information of all trajectory points to finally obtain the environmental attribute error of the matching trajectory segment.

[0092] From the environmental attribute errors of all matching trajectory segments, k (for example, k is 3) matching trajectory segments with the smallest values ​​are selected as candidate trajectory segments.

[0093] Step 104: precisely match the driving trajectory segment with the k candidate trajectory segments, and determine a target trajectory segment from the k candidate trajectory segments to determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

[0094] After obtaining the rough matching result, it is necessary to determine the one that best matches the driving trajectory segment from the above k candidate trajectory segments as the target trajectory segment. The specific precise matching steps may include:

[0095] For each candidate trajectory segment, respectively calculating a total error between the candidate trajectory segment and the driving trajectory segment;

[0096] Determine a target trajectory segment matching the driving trajectory segment from the k candidate trajectory segments according to a total error between each candidate trajectory segment and the driving trajectory segment;

[0097] Get the recursive posture information of the vehicle at the current moment;

[0098] Determine the current target posture information of the vehicle according to the posture information corresponding to the mapping trajectory point with the largest sequence number in the target trajectory segment, the recursive posture information and the driving posture information of the vehicle; wherein the driving posture information is the posture information obtained when the vehicle starts to perform rough matching, the recursive posture information is the posture information at the current moment derived based on the driving posture information, and the target posture information is the posture information calculated after the vehicle performs fine matching processing;

[0099] The position of the vehicle in the parking map is determined according to the target posture information.

[0100] In this embodiment, to calculate the total error between the candidate trajectory segment and the driving trajectory segment, it is necessary to first calculate the driving center point of the driving trajectory segment; the x value of the driving center point is the average of the x values ​​of each driving trajectory point, the y value of the driving center point is the average of the y values ​​of each driving trajectory point, and the z value of the driving center point is the average of the z values ​​of each driving trajectory point. Similarly, the candidate center point of the candidate trajectory segment is calculated.

[0101] Then, for each driving trajectory point in the driving trajectory segment, the target driving trajectory point corresponding to the driving trajectory point is calculated according to the driving center point; that is, the coordinates of each driving trajectory point are subtracted from the coordinates of the driving center point to obtain the target driving trajectory point. Similarly, for each mapping trajectory point in the candidate trajectory segment, the target mapping trajectory point corresponding to the mapping trajectory point is calculated according to the candidate center point.

[0102] Then, according to the target driving trajectory point and the target mapping trajectory point, the residual function of the candidate trajectory segment and the driving trajectory segment about rotation and translation is constructed, and the transformation matrix from the driving trajectory segment to the candidate trajectory segment is calculated. For details, please refer to the following formula:

[0103]

[0104]

[0105] Among them, T op is the transformation matrix, R is the rotation variable, t is the translation variable, e(R, t) is the residual function, N is the number of driving trajectory points in the driving trajectory segment, i ranges from 0 to N, C mi Map the trajectory point of the i-th target in the candidate trajectory segment, C si is the i-th target driving trajectory point in the driving trajectory segment, and tr(*) is the trace of the matrix.

[0106] After the above optimization and derivation, let Perform SVD decomposition (singular value decomposition) on H and get H = UΣV. According to the characteristics of the matrix trace, we know that R = VU T When , the error is the smallest, but it cannot be guaranteed that the modulus of R is 1 at this time, so R needs to be multiplied by a coefficient W, that is, R = VWU T :

[0107]

[0108] The final transformation matrix

[0109] Using the transformation matrix T obtained above op , rotate the driving trajectory segment to the vicinity of the candidate trajectory segment, obtain the rotation trajectory segment of the driving trajectory segment about the candidate trajectory segment, and solve the total error of the one-to-one corresponding trajectory points .in, is the position information of the i-th mapping trajectory point in the candidate trajectory segment, Temp pi is the position of the i-th driving trajectory point in the driving trajectory segment, U is the left singular vector matrix, V is the right singular vector matrix, U s is the driving center point of the driving trajectory segment, U m is the candidate center point of the candidate trajectory segment.

[0110] After calculating the total error for each candidate trajectory segment, the candidate trajectory segment with the smallest total error value is determined as the target trajectory segment.

[0111] It should be noted that both the rough matching and the final matching require processing time. At this time, the vehicle may travel to a new position. In order to synchronize the vehicle position with time, this embodiment will obtain the recursive posture information of the vehicle at the current moment.

[0112] Then, according to the posture information corresponding to the mapping trajectory point with the largest sequence number in the target trajectory segment, the recursive posture information and the driving posture information of the vehicle, the current target posture information of the vehicle is determined; wherein the driving posture information is the posture information obtained when the vehicle starts to perform rough matching, the recursive posture information is the posture information at the current moment derived based on the driving posture information, and the target posture information is the posture information calculated after the vehicle performs fine matching processing;

[0113] Among them, the last mapping trajectory point of the target trajectory segment is the posture information of the vehicle obtained by precise matching, which is converted into a matrix form as T c , the vehicle posture with error is T e , using the previously acquired driving posture information T 0 Finally, the vehicle’s current target position information is calculated as T copt =T c T 0 -1 T e .

[0114] Finally, based on the target posture information, the position of the vehicle in the parking map can be determined, thereby facilitating the online use of the parking function.

[0115] In summary, the parking trajectory matching method provided in the embodiment of this specification obtains the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles; encodes and stores the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles to obtain the driving trajectory data set corresponding to the driving trajectory segment of the vehicle; according to the driving trajectory data set corresponding to the driving trajectory segment, the driving trajectory segment is roughly matched with the mapping trajectory points in the parking map pre-stored in the vehicle, and k candidate trajectory segments are determined from the mapping trajectory points of the parking map, each candidate trajectory segment is composed of N continuous mapping trajectory points; the driving trajectory segment is precisely matched with the k candidate trajectory segments, and the target trajectory segment is determined from the k candidate trajectory segments to determine the position of the vehicle in the parking map. In this way, the rough matching strategy of the trajectory point attributes is used to reduce the probability of mismatching, greatly improve the reliability of the trajectory matching results, and finally the precise matching strategy is used to not only improve the accuracy of trajectory matching, but also improve the matching effect, ensure the success rate of parking trajectory matching of vehicles in underground parking scenarios, and improve user experience.

[0116] Based on the same inventive concept, combined Figure 3As shown, an embodiment of the present invention further provides a parking trajectory matching device, comprising:

[0117] An acquisition module, used to acquire the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles;

[0118] An encoding module, used for encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle, to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle;

[0119] a coarse matching module, configured to coarsely match the driving trajectory segment with the mapping trajectory points in the parking map pre-stored in the vehicle according to the driving trajectory data set corresponding to the driving trajectory segment, and determine k candidate trajectory segments from the mapping trajectory points in the parking map, each candidate trajectory segment consisting of N consecutive mapping trajectory points;

[0120] A precise matching module is used to precisely match the driving trajectory segment with the k candidate trajectory segments, determine a target trajectory segment from the k candidate trajectory segments, and determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

[0121] Optionally, the encoding module is also used to:

[0122] For each driving trajectory point, attribute encoding is performed on the environmental feature corresponding to the driving trajectory point to obtain the environmental attribute information corresponding to the driving trajectory point;

[0123] The driving position and the environmental attribute information corresponding to the driving trajectory point are used as four-dimensional driving trajectory data, wherein N groups of the driving trajectory data constitute the driving trajectory data set;

[0124] Performing validity detection on the driving trajectory data in the driving trajectory data set;

[0125] If the number of valid driving trajectory data in the driving trajectory data set is greater than the number threshold, a rough matching process is performed.

[0126] Optionally, the coarse matching module is also used to:

[0127] Obtain the vehicle's driving posture information;

[0128] Putting the mapping trajectory data corresponding to the mapping trajectory points in the parking map into KDTree, and searching for the mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle;

[0129] If a mapping trajectory point in the parking map is searched within the target radius, a matching trajectory point set is determined according to the sequence number of the searched mapping trajectory point;

[0130] In the matching trajectory point set, starting from the matching trajectory point with the smallest serial number, the matching trajectory point is used as the starting trajectory point, and extraction processing is performed to obtain m matching trajectory segments, and the extraction processing step includes: starting from the starting trajectory point, sequentially extracting N matching trajectory points as a matching trajectory segment; m is an integer greater than 1;

[0131] Traversing all matching trajectory segments, and respectively calculating the environmental attribute error between each matching trajectory segment and the driving trajectory segment;

[0132] According to the environmental attribute errors of the matching trajectory segments, k candidate trajectory segments are determined from the matching trajectory segments.

[0133] Optionally, the coarse matching module is also used to:

[0134] If no mapping trajectory point in the parking map is found within the target radius, the target radius is expanded and updated to obtain an updated target radius, and the step of searching for the mapping trajectory point in the parking map within the target radius is returned to be executed; or,

[0135] If no mapping trajectory point in the parking map is found within the target radius, the process returns to the step of obtaining the driving positions and environmental features corresponding to the driving trajectory points of the N groups of vehicles.

[0136] Optionally, the fine matching module is also used to:

[0137] For each candidate trajectory segment, respectively calculating a total error between the candidate trajectory segment and the driving trajectory segment;

[0138] Determine a target trajectory segment matching the driving trajectory segment from the k candidate trajectory segments according to a total error between each candidate trajectory segment and the driving trajectory segment;

[0139] Get the recursive posture information of the vehicle at the current moment;

[0140] Determine the current target posture information of the vehicle according to the posture information corresponding to the mapping trajectory point with the largest sequence number in the target trajectory segment, the recursive posture information and the driving posture information of the vehicle; wherein the driving posture information is the posture information obtained when the vehicle starts to perform rough matching, the recursive posture information is the posture information at the current moment derived based on the driving posture information, and the target posture information is the posture information calculated after the vehicle performs fine matching processing;

[0141] The position of the vehicle in the parking map is determined according to the target posture information.

[0142] Optionally, the fine matching module is also used to:

[0143] Calculating the driving center point of the driving trajectory segment;

[0144] For each driving trajectory point in the driving trajectory segment, according to the driving center point, calculate a target driving trajectory point corresponding to the driving trajectory point;

[0145] Calculate the candidate center point of the candidate trajectory segment;

[0146] For each mapping trajectory point in the candidate trajectory segment, calculate the target mapping trajectory point corresponding to the mapping trajectory point according to the candidate center point;

[0147] According to the target driving trajectory point and the target mapping trajectory point, a residual function of the candidate trajectory segment and the driving trajectory segment with respect to rotation and translation is constructed, and a transformation matrix from the driving trajectory segment to the candidate trajectory segment is calculated;

[0148] According to the transformation matrix, the driving trajectory segment is rotated and transformed to obtain a rotation trajectory segment of the driving trajectory segment with respect to the candidate trajectory segment;

[0149] The total error of the driving trajectory segment with respect to the candidate trajectory segment is calculated according to the rotation trajectory segment and the candidate trajectory segment.

[0150] Optionally, the acquisition module is also used to:

[0151] Get the vehicle's location;

[0152] If the position of the vehicle is within a preset distance of the area where the parking garage is located, multiple groups of vehicle positions and environmental features corresponding to the trajectory points of the vehicle's travel are obtained; wherein a parking map corresponding to the parking garage is pre-stored in the vehicle.

[0153] In summary, the parking trajectory matching device provided in the embodiment of the specification obtains the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles; encodes and stores the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles to obtain the driving trajectory data set corresponding to the driving trajectory segment of the vehicle; according to the driving trajectory data set corresponding to the driving trajectory segment, the driving trajectory segment is roughly matched with the mapping trajectory points in the parking map pre-stored in the vehicle, and k candidate trajectory segments are determined from the mapping trajectory points of the parking map, each candidate trajectory segment is composed of N continuous mapping trajectory points; the driving trajectory segment is precisely matched with the k candidate trajectory segments, and the target trajectory segment is determined from the k candidate trajectory segments to determine the position of the vehicle in the parking map. In this way, the rough matching strategy of the trajectory point attributes is used to reduce the probability of mismatching, greatly improve the reliability of the trajectory matching results, and finally the precise matching strategy is used to not only improve the accuracy of trajectory matching, but also improve the matching effect, ensure the success rate of parking trajectory matching of vehicles in underground parking scenarios, and improve user experience.

[0154] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the parking trajectory matching device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

[0155] Based on the same inventive concept, an embodiment of the present invention further provides a controller, which includes a parking trajectory matching device, a memory, a processor and a communication unit. The memory stores machine-readable instructions executable by the processor. When the controller is running, the processor and the memory communicate through a bus, the processor executes the machine-readable instructions, and executes the parking trajectory matching method.

[0156] The memory, the processor, and the communication unit are electrically connected to each other directly or indirectly to achieve signal transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines. The parking trajectory matching device includes at least one software function module that can be stored in the memory in the form of software or firmware. The processor is used to execute the executable module stored in the memory (for example, the software function module or computer program included in the parking trajectory matching device).

[0157] Among them, the memory can be, but is not limited to, random access memory (Random Access Memory, RAM), read only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.

[0158] In some embodiments, the processor is used to perform one or more functions described in this embodiment. In some embodiments, the processor may include one or more processing cores (eg, a single-core processor (S) or a multi-core processor (S)).

[0159] In this embodiment, the memory is used to store the program, and the processor is used to execute the program after receiving the execution instruction. The process definition method disclosed in any implementation of this embodiment can be applied to the processor, or implemented by the processor.

[0160] The communication unit is used to establish a communication connection between the controller and other devices through the network, and to send and receive data through the network.

[0161] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the controller described above can refer to the corresponding process in the aforementioned method, and will not be elaborated here.

[0162] Based on the same inventive concept, an embodiment of the present invention further provides a vehicle, including a vehicle body and a controller installed in the vehicle body, wherein the controller is used to implement the aforementioned parking trajectory matching method.

[0163] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the controller of the vehicle described above can refer to the corresponding process in the aforementioned method, and will not be elaborated here.

[0164] The above are only various embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A parking trajectory matching method, characterized in that: include: Obtain the driving positions and environmental characteristics corresponding to the driving trajectory points of N groups of vehicles; Encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle; According to the driving trajectory data set corresponding to the driving trajectory segment, the driving trajectory segment is roughly matched with the mapping trajectory points in the parking map pre-stored in the vehicle, and k candidate trajectory segments are determined from the mapping trajectory points in the parking map, each candidate trajectory segment is composed of N continuous mapping trajectory points; The driving trajectory segment is precisely matched with the k candidate trajectory segments, and a target trajectory segment is determined from the k candidate trajectory segments to determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

2. The method according to claim 1, characterized in that: The encoding and storage processing of the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle includes: For each driving trajectory point, attribute encoding is performed on the environmental feature corresponding to the driving trajectory point to obtain the environmental attribute information corresponding to the driving trajectory point; The driving position and the environmental attribute information corresponding to the driving trajectory point are used as four-dimensional driving trajectory data, wherein N groups of the driving trajectory data constitute the driving trajectory data set; Performing validity detection on the driving trajectory data in the driving trajectory data set; If the number of valid driving trajectory data in the driving trajectory data set is greater than the number threshold, a rough matching process is performed.

3. The method according to claim 1, characterized in that The method of roughly matching the driving trajectory segment with a mapping trajectory point in a parking map pre-stored in the vehicle according to the driving trajectory data set corresponding to the driving trajectory segment, and determining k candidate trajectory segments from the mapping trajectory points in the parking map, comprises: Obtain the vehicle's driving posture information; Putting the mapping trajectory data corresponding to the mapping trajectory points in the parking map into KDTree, and searching for the mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle; If a mapping trajectory point in the parking map is searched within the target radius, a matching trajectory point set is determined according to the sequence number of the searched mapping trajectory point; In the matching trajectory point set, starting from the matching trajectory point with the smallest serial number, the matching trajectory point is used as the starting trajectory point, and extraction processing is performed to obtain m matching trajectory segments, and the extraction processing step includes: starting from the starting trajectory point, sequentially extracting N matching trajectory points as a matching trajectory segment; m is an integer greater than 1; Traversing all matching trajectory segments, and respectively calculating the environmental attribute error between each matching trajectory segment and the driving trajectory segment; According to the environmental attribute errors of the matching trajectory segments, k candidate trajectory segments are determined from the matching trajectory segments.

4. The method according to claim 3, characterized in that After the step of searching for mapping trajectory points in the parking map within a target radius according to the driving posture information of the vehicle, the method further includes: If no mapping trajectory point in the parking map is found within the target radius, the target radius is expanded and updated to obtain an updated target radius, and the step of searching for the mapping trajectory point in the parking map within the target radius is returned to be executed; or, If no mapping trajectory point in the parking map is found within the target radius, the process returns to the step of obtaining the driving positions and environmental features corresponding to the driving trajectory points of the N groups of vehicles.

5. The method according to claim 1, characterized in that The step of precisely matching the driving trajectory segment with the k candidate trajectory segments and determining a target trajectory segment from the k candidate trajectory segments to determine the position of the vehicle in the parking map includes: For each candidate trajectory segment, respectively calculating a total error between the candidate trajectory segment and the driving trajectory segment; Determine a target trajectory segment matching the driving trajectory segment from the k candidate trajectory segments according to a total error between each candidate trajectory segment and the driving trajectory segment; Get the recursive posture information of the vehicle at the current moment; Determine the current target posture information of the vehicle according to the posture information corresponding to the mapping trajectory point with the largest sequence number in the target trajectory segment, the recursive posture information and the driving posture information of the vehicle; wherein the driving posture information is the posture information obtained when the vehicle starts to perform rough matching, the recursive posture information is the posture information at the current moment derived based on the driving posture information, and the target posture information is the posture information calculated after the vehicle performs fine matching processing; The position of the vehicle in the parking map is determined according to the target posture information.

6. The method according to claim 5, characterized in that The step of calculating, for each candidate trajectory segment, a total error between the candidate trajectory segment and the driving trajectory segment includes: Calculating the driving center point of the driving trajectory segment; For each driving trajectory point in the driving trajectory segment, according to the driving center point, calculate a target driving trajectory point corresponding to the driving trajectory point; Calculate the candidate center point of the candidate trajectory segment; For each mapping trajectory point in the candidate trajectory segment, calculate the target mapping trajectory point corresponding to the mapping trajectory point according to the candidate center point; According to the target driving trajectory point and the target mapping trajectory point, a residual function of the candidate trajectory segment and the driving trajectory segment with respect to rotation and translation is constructed, and a transformation matrix from the driving trajectory segment to the candidate trajectory segment is calculated; According to the transformation matrix, the driving trajectory segment is rotated and transformed to obtain a rotation trajectory segment of the driving trajectory segment with respect to the candidate trajectory segment; The total error of the driving trajectory segment with respect to the candidate trajectory segment is calculated according to the rotation trajectory segment and the candidate trajectory segment.

7. The method according to claim 1, characterized in that The step of obtaining the positions of multiple groups of vehicles and the environmental features corresponding to the trajectory points of the vehicles includes: Get the vehicle's location; If the position of the vehicle is within a preset distance of the area where the parking garage is located, multiple groups of vehicle positions and environmental features corresponding to the trajectory points of the vehicle's travel are obtained; wherein a parking map corresponding to the parking garage is pre-stored in the vehicle.

8. A parking trajectory matching device, characterized in that: include: An acquisition module, used to acquire the driving positions and environmental features corresponding to the driving trajectory points of N groups of vehicles; An encoding module, used for encoding and storing the driving positions and environmental features corresponding to the N groups of driving trajectory points of the vehicle, to obtain a driving trajectory data set corresponding to the driving trajectory segment of the vehicle; a coarse matching module, configured to coarsely match the driving trajectory segment with the mapping trajectory points in the parking map pre-stored in the vehicle according to the driving trajectory data set corresponding to the driving trajectory segment, and determine k candidate trajectory segments from the mapping trajectory points in the parking map, each candidate trajectory segment consisting of N consecutive mapping trajectory points; A precise matching module is used to precisely match the driving trajectory segment with the k candidate trajectory segments, determine a target trajectory segment from the k candidate trajectory segments, and determine the position of the vehicle in the parking map, wherein k and N are both integers greater than 1.

9. A controller, characterized in that: The controller includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the parking trajectory matching method described in any one of claims 1 to 7 is implemented.

10. A vehicle, characterized in that: The vehicle comprises a vehicle body and a controller installed in the vehicle body, wherein the controller executes the parking trajectory matching method according to any one of claims 1-7.

Images