Vehicle pose positioning method, system, device and storage medium

Abstract

The application relates to a vehicle pose positioning method, system, device and storage medium, and the technical scheme points are as follows: vehicle position data and laser point cloud data are acquired; the laser point cloud data is filtered according to the vehicle position data to obtain a first vehicle laser point set, and a vehicle positioning frame is obtained according to the first vehicle laser point set; a plurality of wheel samples are obtained according to the vehicle positioning frame, a preset wheelbase range and preset segmentation data; the first vehicle laser point set is converted into a template coordinate system according to a preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set; each second vehicle laser point set is compared with a corresponding wheel template to obtain a comparison result, and a target vehicle laser point set is selected from all the second vehicle laser point sets according to all the comparison results; the target vehicle laser point set is subjected to point cloud classification and error minimization fitting according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose; the application has the effects that the vehicle is positioned by using a wheelbase and a wheelbase center, the positioning is accurate and fast, and the positioning speed of the vehicle is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle positioning, more particularly, it relates to a vehicle pose positioning method, system, device and storage medium. BACKGROUND

[0002] At present, in the process of transporting vehicles, a vehicle transporting robot is usually used to transport vehicles, such as a vehicle transporting robot of a vehicle loading plate type, a comb type or a clamping wheel type.

[0003] The document with the publication number CN110794417A and the name "Method for realizing automobile pose measurement and positioning based on STM32 and laser radar" discloses that after the STM32 vehicle-mounted controller receives the measurement information of the first laser radar and the second laser radar, a vehicle body point model under a coordinate system is established, n pairs of effective data points on the left and right sides are selected in the coordinate system and are subjected to mean value processing, and a point line composed of all the mean value points is the center line of the laser radar scanning vehicle body model. The relative position relationship between the center point of the vehicle to be transported and the center point of the transporting AGV can be calculated by fitting the mean value points, and then the transporting AGV is controlled to adjust the pose. However, the vehicle body pose is determined by establishing a vehicle body point model and then selecting n pairs of effective data points on the left and right sides, and a large number of effective data points are required, which leads to a longer time required for the process of obtaining the center line of the vehicle body model from the effective data points. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application aims to provide a vehicle pose positioning method, system, device and storage medium, which has the functional advantages of positioning the vehicle by using the wheel track and the wheel track center, and being accurate and fast in positioning, thereby improving the positioning speed of the vehicle.

[0005] The above technical purpose of the present application is achieved by the following technical scheme:

[0006] A vehicle pose positioning method comprises the following steps:

[0007] S1, acquiring parking space data and laser point cloud data containing vehicle wheels in a vehicle, wherein the parking space data is the parking space position data of a parking space where the vehicle is parked;

[0008] S2, filtering the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtaining a vehicle positioning frame according to the first vehicle laser point set;

[0009] S3, obtaining a plurality of wheel samples according to the vehicle positioning frame, a preset wheel track range and preset segmentation data;

[0010] S4, converting the first vehicle laser point set into a template coordinate system according to the preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval;

[0011] S5, determining the corresponding second vehicle laser point set according to each wheel sample, and determining the corresponding wheel template from the wheel template set, comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results;

[0012] S6, performing point cloud classification and error minimization fitting on the target vehicle laser point set according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose.

[0013] Optionally, the parking space data is obtained from a parking lot coordinate system constructed by a SLAM method; the vehicle laser data points are generated based on a laser radar coordinate system; and the S2 includes the following sub-steps:

[0014] S21, calibrating the external parameters of a parking lot and a laser radar to determine the correspondence relationship between the parking lot and the laser radar, and converting the parking space data into a laser radar coordinate system to obtain parking space laser data according to the correspondence relationship;

[0015] S22, comparing the parking space laser data with the laser point cloud data to filter out laser point cloud data located in the parking space to obtain a plurality of laser data points to form a parking space laser point set;

[0016] S23, reducing the parking space laser data according to a preset ratio to obtain reduced parking space laser data, and filtering the parking space laser point set according to the reduced parking space laser data to obtain a plurality of filtered laser data points to form a filtered laser point set;

[0017] S24, determining whether the difference between the number of laser data points of the parking space laser point set and the filtered laser point set is greater than a preset threshold, if yes, taking the parking space laser point set as the first vehicle laser point set, and performing sub-step S25, if not, taking the reduced parking space laser data as the parking space laser data, taking the filtered laser point set as the parking space laser point set, and returning to perform sub-step S23;

[0018] S25, calculating the first vehicle laser point set by using a minimum bounding box algorithm to obtain a vehicle positioning frame.

[0019] Optionally, the preset segmentation data comprises: a first segmentation number in the x dimension of the laser radar coordinate system, a second segmentation number in the y dimension of the laser radar coordinate system, a third segmentation number in the theta dimension of the laser radar coordinate system, and a fourth segmentation number in the preset wheelbase range; and the S3 comprises the following sub-steps:

[0020] S31, segmenting the vehicle positioning frame in the x dimension of the laser radar coordinate system according to the first segmentation number to obtain a plurality of first planes;

[0021] S32, segmenting the vehicle positioning frame in the y dimension of the laser radar coordinate system according to the second segmentation number to obtain a plurality of second planes;

[0022] S33, segmenting the vehicle positioning frame in the theta dimension of the laser radar coordinate system according to the third segmentation number to obtain a plurality of third planes;

[0023] S34, obtaining a plurality of intersection points according to the intersection of the first plane, the second plane and the third plane;

[0024] S35, selecting a plurality of sample wheelbases in the preset wheelbase range according to the fourth segmentation number;

[0025] S36, combining each sample wheelbase with each intersection point to obtain a corresponding wheel sample.

[0026] Optionally, the S4 comprises the following steps:

[0027] S41, obtaining a first positioning coordinate of a car-moving robot in a world coordinate system and a second positioning coordinate of a laser radar in a car-moving robot coordinate system, and converting a first vehicle laser point set to the car-moving robot coordinate system to obtain a first sub-vehicle laser point set according to the first positioning coordinate, the second positioning coordinate and a preset first conversion formula, wherein the first conversion formula is, ,

[0028] wherein xm represents a value in the X dimension of the car-moving robot coordinate system, ym represents a value in the Y dimension of the car-moving robot coordinate system, xR represents a value of the X dimension of the car-moving robot in the world coordinate system, yR represents a value of the X dimension of the car-moving robot in the world coordinate system, thetaR represents a value of the theta dimension of the car-moving robot in the world coordinate system, xL represents a value of the X dimension of the laser radar in the car-moving robot coordinate system, yL represents a value of the Y dimension of the laser radar in the car-moving robot coordinate system, thetaL represents a value of the theta dimension of the laser radar in the car-moving robot coordinate system, xP represents a value in the X dimension of the laser radar coordinate system, and yP represents a value in the Y dimension of the laser radar coordinate system.

[0029] S42、According to the preset second conversion formula and each wheel sample, the first vehicle laser point set is converted into a template coordinate system to obtain a second vehicle laser point set, and the second conversion formula is,

[0030] Wherein, xw represents a value in the X dimension of the template coordinate system, yw represents a value in the Y dimension of the template coordinate system, xw1 represents a value of the intersection point in the wheel sample in the X dimension of the robot coordinate system, yw1 represents a value of the intersection point in the wheel sample in the Y dimension of the robot coordinate system, and thetaw1 represents a value of the intersection point in the wheel sample in the theta dimension of the robot coordinate system.

[0031] Optionally, the wheel template comprises a wheel area and a spacing area; the comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results, comprises:

[0032] Comparing each second vehicle laser point set with the corresponding wheel template to judge whether each data laser point in the second vehicle laser point set is located in the wheel area in the corresponding wheel template, if yes, it is recorded as 1 point, if not, it is recorded as a to-be-compared data laser point, and it is judged whether the to-be-compared data laser point is located in the spacing area in the corresponding wheel template, if yes, it is recorded as-1 point, if not, it is recorded as 0 point;

[0033] Adding up the scores of all data laser points in each second vehicle laser point set to obtain the corresponding comparison result;

[0034] Comparing the comparison results of all second vehicle laser point sets, and selecting the second vehicle laser point set with the highest score as the target vehicle laser point set.

[0035] Optionally, the wheel template is divided into a tire surface area, a first tire area and a second tire area; S6 comprises the following sub-steps:

[0036] Comparing the target vehicle laser point set with the corresponding wheel template to judge whether each data laser point in the target vehicle laser point set is located in the tire surface area in the corresponding wheel template, if yes, it is recorded as a first data point, if not, it is recorded as a first to-be-judged data point, and it is judged whether the first to-be-judged data point is located in the first tire area in the corresponding wheel template, if yes, it is recorded as a second data point, if not, it is recorded as a second to-be-judged data point, and it is judged whether the second to-be-judged data point is located in the second tire area in the corresponding wheel template, if yes, it is recorded as a third data point, if not, it is recorded as an invalid data point.

[0037] All the first data points form a first data point set, all the second data points form a second data point set, and all the third data points form a third data point set, and an error minimization problem is established according to the first data point set, the second data point set and the third data point set as follows:

[0038] Wherein, X(k), Y(k) represent the coordinates of the data points in the first data point set, X(i), Y(i) represent the coordinates of the data points in the second data point set, X(j), Y(j) represent the coordinates of the data points in the third data point set, L1 represents the first data point set, L2 represents the second data point set, and L3 represents the third data point set, and the values of A, B, C and D are obtained by solving the error minimization problem by using a convex optimization method;

[0039] The values of A and D are taken as parameters of the first straight line equation to obtain the first straight line, the values of A and B are taken as parameters of the second straight line equation to obtain the second straight line, and the values of A and C are taken as parameters of the third straight line equation to obtain the third straight line;

[0040] The intersection of the first straight line and the second straight line is taken as the left wheel position of the vehicle, the intersection of the first straight line and the third straight line is taken as the right wheel position of the vehicle, and the slope of the first straight line is taken as the wheel posture of the vehicle.

[0041] Optionally, after the step S6, the following steps are further included:

[0042] A first positioning coordinate of the vehicle moving robot in a world coordinate system is obtained;

[0043] The left wheel position is converted into the world coordinate system according to a preset third conversion formula, the first positioning coordinate and a target wheel sample to obtain an actual left wheel position, wherein the target wheel sample is a wheel sample corresponding to a target vehicle laser point set;

[0044] The right wheel position is converted into the world coordinate system according to the preset third conversion formula, the first positioning coordinate and the target wheel sample to obtain an actual right wheel position;

[0045] The third conversion formula is as follows:

[0046] Wherein, x represents the value of the actual left wheel position or the actual right wheel position in the X dimension of the world coordinate system, y represents the value of the actual left wheel position or the actual right wheel position in the Y dimension of the world coordinate system, xw2 represents the value of the target wheel sample in the X dimension of the robot coordinate system, yw2 represents the value of the target wheel sample in the Y dimension of the robot coordinate system, thetaw2 represents the value of the target wheel sample in the theta dimension of the robot coordinate system, xR represents the value of the robot in the X dimension of the world coordinate system, yR represents the value of the robot in the X dimension of the world coordinate system, and thetaR represents the value of the robot in the theta dimension of the world coordinate system, xLFS represents the value of the left wheel position or the right wheel position in the X dimension of the template coordinate system, and yLFS represents the value of the left wheel position or the right wheel position in the Y dimension of the template coordinate system.

[0047] A vehicle pose positioning system, comprising:

[0048] A data acquisition module is configured to acquire parking space data and laser point cloud data containing wheels in a vehicle, wherein the parking space data is parking space position data of a parking space where the vehicle is parked.

[0049] A data filtering module is configured to filter the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtain a vehicle positioning frame according to the first vehicle laser point set.

[0050] A sample determination module is configured to obtain a plurality of wheel samples according to the vehicle positioning frame, a preset wheelbase range, and preset segmentation data.

[0051] A coordinate conversion module is configured to convert the first vehicle laser point set to a template coordinate system according to a preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval.

[0052] A data comparison module is configured to determine the corresponding second vehicle laser point set according to each wheel sample, determine the corresponding wheel template from the wheel template set, compare each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and select a target vehicle laser point set from all second vehicle laser point sets according to all comparison results.

[0053] A pose determination module is configured to perform point cloud classification and error minimization fitting on the target vehicle laser point set according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose.

[0054] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method when executing the computer program.

[0055] A computer readable storage medium, which stores a computer program, the computer program is executed by a processor to implement the steps of the method.

[0056] In summary, the present application has the following beneficial effects: first, the first vehicle laser point set is obtained by filtering the laser point cloud data through the parking space data, the vehicle positioning frame is obtained according to the first vehicle laser point set, the detection range is greatly reduced, the vehicle positioning speed is improved, the vehicle positioning frame is segmented to obtain a plurality of wheel samples, the laser data points in the first laser point set are converted according to each wheel sample to obtain the corresponding second vehicle laser point set, each second vehicle laser point set is compared with the corresponding wheel template, the second vehicle laser point set with the highest similarity of the comparison result is taken as the target vehicle laser point set, the wheel template corresponding to the target vehicle laser point set is the target wheel template, and the left wheel position, the right wheel position and the wheel posture of the vehicle are obtained by point cloud classification and error minimization fitting of the target vehicle laser point set according to the target wheel template, that is, the target vehicle pose is obtained, the vehicle is handled by the vehicle moving robot, the vehicle is mainly positioned by the wheel track and the wheel track center, the positioning is accurate and fast, and the positioning speed of the vehicle is improved. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 It is a flowchart of the vehicle pose positioning method provided by the present application;

[0058] Figure 2 It is a comparison diagram of the second vehicle laser point set and the wheel template in the present application;

[0059] Figure 3 It is a structure block diagram of the vehicle pose positioning system provided by the present application;

[0060] Figure 4 It is an internal structure diagram of the computer device in the embodiment of the present application.

[0061] In the figure: 1, wheel area; 2, interval area; 3, invalid area; 4, target vehicle laser point set; 5, tire surface area; 6, first tire area; 7, second tire area; 8, origin of template coordinate system. DETAILED DESCRIPTION

[0062] In order to make the objects, characteristics and advantages of the present application more apparent, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. The present application is illustrated by several embodiments in the drawings. However, the present application can be realized in many different forms and is not limited to the embodiments described herein.

[0063] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting", "fixing" and the like should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. The terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more features.

[0064] The present application will be described in detail below with reference to the accompanying drawings and examples.

[0065] The present application provides a vehicle pose positioning method, as shown in Figure 1 The present application provides a vehicle pose positioning method, as shown in

[0066] S1, acquiring parking space data and laser point cloud data containing wheels in the vehicle, the parking space data being parking space position data of a parking space where the vehicle is parked;

[0067] S2, filtering the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtaining a vehicle positioning frame according to the first vehicle laser point set;

[0068] S3, obtaining a plurality of wheel samples according to the vehicle positioning frame, a preset wheelbase range and a preset segmentation data;

[0069] S4, converting the first vehicle laser point set into a template coordinate system according to a preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval;

[0070] S5, determining the corresponding second vehicle laser point set according to each wheel sample, and determining the corresponding wheel template from the wheel template set, comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results;

[0071] S6, performing point cloud classification and error minimization fitting on the target vehicle laser point set according to the wheel template corresponding to the target vehicle laser point set to obtain a target vehicle pose.

[0072] In actual application, since the positions of each parking space are determined when the parking lot is established, the specific position of the vehicle can be determined according to the GPS information of the vehicle, so that the specific parking space where the vehicle is parked in the parking lot can be determined according to the specific position of the vehicle, and the position data of the parking space is obtained, that is, the parking space data is obtained. The laser point cloud data is obtained by scanning the vehicle once by the laser radar on the vehicle moving robot. The laser radar can be a 16-line laser radar. When parking, the vehicle is usually located in the parking space. Therefore, the laser data points not belonging to the vehicle can be filtered out by filtering the laser point cloud data according to the parking space data, so as to obtain the laser data points for representing the approximate position of the vehicle to form a first vehicle laser point set. The vehicle positioning frame is obtained according to the first vehicle laser point set. The vehicle positioning frame is segmented to form a plurality of wheel samples with different wheelbases by using a preset wheelbase range and a preset segmentation data. Then, the laser data points in the first laser point set are converted to obtain a corresponding second vehicle laser point set according to the corresponding relationship between each wheel sample and the corresponding wheel template. Then, each second vehicle laser point set is compared with the corresponding wheel template to obtain a comparison result. The second vehicle laser point set with the highest similarity of the comparison result is taken as the target vehicle laser point set, that is, the wheelbase of the vehicle is determined. The wheelbase of the vehicle is denoted as a target wheelbase. The wheel template corresponding to the target vehicle laser point set is denoted as a target wheel template. The target wheel template is the most suitable wheel template for the vehicle. The left wheel position, the right wheel position and the wheel pose of the vehicle are obtained by performing point cloud classification and error minimization fitting on the target vehicle laser point set according to the target wheel template, that is, the target vehicle pose is obtained, so as to facilitate the vehicle moving robot to perform vehicle moving processing on the vehicle.

[0073] Further, the parking space data is obtained from a parking lot coordinate system constructed by a SLAM method; the vehicle laser data points are generated based on a laser radar coordinate system; and the S2 comprises the following sub-steps:

[0074] S21, performing external parameter calibration on the parking lot and the laser radar to determine the corresponding relationship between the parking lot and the laser radar, and converting the parking space data into the laser radar coordinate system to obtain parking space laser data according to the corresponding relationship;

[0075] S22, comparing the parking space laser data with the laser point cloud data to filter out the laser point cloud data located in the parking space to obtain a plurality of laser data points to form a parking space laser point set;

[0076] S23, according to the preset ratio to the parking laser data obtained by reducing the reduced parking laser data, according to the reduced parking laser data on the parking laser point set filtering to obtain a plurality of filtered laser data points, forming a filtered laser point set;

[0077] S24, judge the difference between the number of laser data points of the parking laser point set and the filtered laser point set is greater than the preset threshold, if yes, the parking laser point set as the first vehicle laser point set, and execute sub step S25, if not, the reduced parking laser data as the parking laser data, the filtered laser point set as the parking laser point set, and return to execute sub step S23;

[0078] S25, using the minimum bounding box algorithm for calculating the first vehicle laser point set obtained by vehicle positioning frame.

[0079] In practical applications, a parking lot coordinate system is constructed by a SLAM (Simultaneous Localization and Mapping) method, position data of a parking space where a vehicle is parked in the parking lot coordinate system is determined according to GPS information of the vehicle, that is, parking space data is obtained, parking space laser data is obtained by converting the parking space data to the laser radar coordinate system, the parking space laser data is compared with the laser point cloud data, that is, each laser data point in the laser point cloud data is compared with the parking space laser data, in a case where values of each laser data point in the laser point cloud data in each dimension (such as X dimension, Y dimension and Z dimension) are located within each dimension interval of the parking space laser data, the laser data point is filtered out, a plurality of laser data points are filtered out to form a parking space laser point set, the parking space laser data is reduced according to a preset ratio to obtain reduced parking space laser data, that is, each dimension interval of the parking space laser data is reduced according to the preset ratio, such as when the parking space laser data in the X dimension interval is (-1.5, 1.5), the Y dimension interval is (-3, 3), the Z dimension interval is (0, 3), and the preset ratio is n, n is a positive number less than 1, the X dimension interval of the parking space laser data after the first reduction is (-1.5n, 1.5n), the Y dimension interval is (-3n, 3n), and the theta dimension interval is (0, 3n), then the parking space laser point set is filtered according to the reduced parking space laser data to obtain a plurality of filtered laser data points to form a filtered laser point set, then it is judged whether the difference between the number of laser data points in the parking space laser point set and the number of laser data points in the filtered laser point set is greater than a preset threshold, that is, it is judged whether there is a mutation in the number of laser data points in the parking space laser point set after the parking space laser data is reduced compared with the number of laser data points in the parking space laser point set before the reduction, if not, the parking space laser data is continuously reduced, the parking space laser point set is filtered using the reduced parking space laser data, and if yes, it indicates that the parking space laser point set filtered before the reduction can approximately represent the pose of the vehicle, the parking space laser point set filtered before the reduction is taken as a first vehicle laser point set, and the position of the vehicle is approximately determined, after the rough vehicle laser point set is obtained, the first vehicle laser point set is processed by using a minimum bounding box algorithm to obtain a vehicle positioning frame, and the position of the vehicle is determined.

[0080] Further, the preset segmentation data includes: a first segmentation number in the x dimension of the laser radar coordinate system, a second segmentation number in the y dimension of the laser radar coordinate system, a third segmentation number in the theta dimension of the laser radar coordinate system, and a fourth segmentation number in the preset wheelbase range; and the S3 includes the following sub-steps:

[0081] S31, segment the vehicle positioning frame in the x dimension of the laser radar coordinate system according to the first segmentation number to obtain a plurality of first planes;

[0082] S32, segment the vehicle positioning frame in the y dimension of the laser radar coordinate system according to the second segmentation number to obtain a plurality of second planes;

[0083] S33, segment the vehicle positioning frame in the theta dimension of the laser radar coordinate system according to the third segmentation number to obtain a plurality of third planes;

[0084] S34, obtain a plurality of intersection points according to the intersection of the first plane, the second plane and the third plane;

[0085] S35, select a plurality of sample wheelbases in a preset wheelbase range according to the fourth segmentation number;

[0086] S36, combine each sample wheelbase with each intersection point in turn to obtain a corresponding wheel sample.

[0087] In actual application, the higher the values of the first segmentation number, the second segmentation number, the third segmentation number and the fourth segmentation number, the higher the precision. The first segmentation number, the second segmentation number, the third segmentation number and the fourth segmentation number can be adjusted according to actual needs. A plurality of intersection points are obtained by segmenting the vehicle positioning frame according to the first segmentation number, the second segmentation number and the third segmentation number. Each intersection point is regarded as the center of the axle closest to the vehicle positioning robot. The preset wheelbase range can be set according to the actual wheelbase range in the prior art. Each intersection point is combined with each sample wheelbase in turn to obtain a corresponding wheel sample. If the intersection point is represented as (xM, yM, thetaM), the sample wheelbase is represented as wheelbase, and the wheel sample can be represented as (xM, yM, thetaM, wheelbase), xM, yM and thetaM represent the values of the intersection point in the X dimension, the Y dimension and the Theta dimension of the laser radar coordinate system, respectively. In actual application, the wheel template set is represented as

[0088] wherein Nwheelbase represents the number of samples, A wheel template representing a wheel track j, thus, j represents both the wheel track and the template index, the wheel template is a likelihood domain model, including a wheel region 1, a gap region 2 and an invalid region 3, the wheel region is determined according to the wheel track in a preset wheel track range, and is used for modeling a wheel shape, the gap region is determined according to a preset gap, and the region other than the wheel region and the gap region in the wheel template is the invalid region, the wheel template is represented in the form of a grid map, the first vehicle laser point set is converted into a template coordinate system to obtain a second vehicle laser point set according to the obtained wheel sample, so as to be compared with the corresponding wheel template, to determine whether each laser data point in the second vehicle laser point set belongs to the wheel region, the gap region or the invalid region in the wheel template, and an comparison result is obtained.

[0089] Further, the S4 comprises the following steps:

[0090] S41, obtaining a first positioning coordinate of the car moving robot in a world coordinate system and a second positioning coordinate of the laser radar in a car moving robot coordinate system, and converting the first vehicle laser point set into the car moving robot coordinate system to obtain a first sub-vehicle laser point set according to the first positioning coordinate, the second positioning coordinate and a preset first conversion formula, wherein the first conversion formula is, ,

[0091] wherein xm represents a value in the X dimension of the car moving robot coordinate system, ym represents a value in the Y dimension of the car moving robot coordinate system, xR represents a value of the X dimension of the car moving robot in the world coordinate system, yR represents a value of the X dimension of the car moving robot in the world coordinate system, thetaR represents a value of the theta dimension of the car moving robot in the world coordinate system, xL represents a value of the X dimension of the laser radar in the car moving robot coordinate system, yL represents a value of the Y dimension of the laser radar in the car moving robot coordinate system, thetaL represents a value of the theta dimension of the laser radar in the car moving robot coordinate system, xP represents a value of the X dimension of the laser radar coordinate system, and yP represents a value of the Y dimension of the laser radar coordinate system;

[0092] S42, converting the first sub-vehicle laser point set into the template coordinate system to obtain the second vehicle laser point set according to a preset second conversion formula and each wheel sample, and the second conversion formula is,

[0093]

[0094] Wherein, xw represents the value in the X dimension of the template coordinate system, yw represents the value in the Y dimension of the template coordinate system, xw1 represents the value of the intersection point in the wheel sample in the X dimension of the palletizing robot coordinate system, yw1 represents the value of the intersection point in the wheel sample in the Y dimension of the palletizing robot coordinate system, thetaw1 represents the value of the intersection point in the wheel sample in the theta dimension of the palletizing robot coordinate system.

[0095] In practical application, (xP, yP) represents the coordinates in the laser radar coordinate system, (xm, ym) represents the coordinates in the palletizing robot coordinate system, (xR, yR, thetaR) represents the first positioning coordinates of the palletizing robot in the world coordinate system, (xL, yL, thetaL) represents the second positioning coordinates of the laser radar in the world coordinate system, the corresponding relationship between the palletizing robot and the laser radar is obtained according to the first positioning coordinates and the second positioning coordinates, and the first vehicle laser point set in the laser radar coordinate system can be converted into the palletizing robot coordinate system according to the corresponding relationship, and the specific conversion formula is the first conversion formula. Each laser data point in the first vehicle laser point set is input into the first conversion formula to obtain the corresponding laser data point in the palletizing robot coordinate system to form the first sub-vehicle laser point set. After obtaining the first sub-vehicle laser point set, the laser data points in the first sub-vehicle laser point set need to be converted into the template coordinate system. Specifically, (xw1, yw1) represents the coordinates in the template coordinate system, (xw1, yw1, thetaw1) represents the coordinates of the intersection point in the wheel sample in the palletizing robot coordinate system. In order to obtain (xw1, yw1, thetaw1), the intersection point in the laser radar coordinate system needs to be converted into the palletizing robot coordinate system according to the first conversion formula to obtain xw1 and yw1. Since the laser radar coordinate system and the palletizing robot coordinate system are translated in the xy plane, thetaw1 and the value of the intersection point in the laser radar coordinate system in the theta dimension are the same. In the conversion process, each wheel sample selects the corresponding wheel template from the wheel template set according to its wheelbase. If there is a wheel template in the wheel template set that does not have the same wheelbase as the wheel sample, the wheel template with the wheelbase closest to the wheelbase of the wheel sample is selected from the wheel template set. The template coordinate system of the wheel template takes the center point of the wheelbase as the origin. The conversion relationship between each wheel sample and the template coordinate system is determined according to the origin 8 of the template coordinate system and the intersection point of each wheel sample. The first sub-vehicle laser point set is converted into the template coordinate system according to each conversion relationship to obtain the corresponding second vehicle laser point set. The specific conversion formula is the second conversion formula.

[0096] Further, the comparing each of the second vehicle laser point sets with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all the second vehicle laser point sets according to all the comparison results, comprises:

[0097] comparing each of the second vehicle laser point sets with the corresponding wheel template, judging whether each data laser point in the second vehicle laser point set is located in the wheel region in the corresponding wheel template, if yes, recording 1 point, if not, recording a to-be-compared data laser point, and judging whether the to-be-compared data laser point is located in the interval region in the corresponding wheel template, if yes, recording-1 point, if not, recording 0 point;

[0098] adding up the scores of all data laser points in each of the second vehicle laser point sets to obtain a corresponding comparison result;

[0099] comparing the comparison results of all the second vehicle laser point sets, and selecting a second vehicle laser point set with the highest score as the target vehicle laser point set.

[0100] In actual application, in the process of comparison, if the laser data point in the second laser point set falls into the wheel region, it means that the laser data point provides positive information, and then records 1 point, if the laser data point in the second laser point set falls into the interval region, it means that the laser data point provides negative information, and then records-1 point, if the laser data point in the second laser point set falls into the invalid region, it means that the laser data point does not provide effective information, and then records-1 point, so as to obtain the score of all laser data points in each second laser point set, and add up the scores of all laser data points in each second laser point set to obtain the comparison score of the second laser point set, that is, the comparison result, the higher the comparison score is, the more matched the corresponding wheel template of the second laser point set is with the vehicle, then the second laser point set with the highest comparison score is selected as the target vehicle laser point set, and the corresponding wheel template of the target vehicle laser point set is the target wheel template, which means that the target wheel template is the most matched template with the vehicle.

[0101] Further, as shown in Figure 2 the wheel template is divided into a tire surface region 5, a first tire region 6 and a second tire region 7; the S6 comprises the following sub-steps:

[0102] The target vehicle laser point set 4 is compared with the corresponding wheel template to determine whether each data laser point in the target vehicle laser point set 4 is located in the tire surface area 5 in the corresponding wheel template, and if so, it is recorded as a first data point, and if not, it is recorded as a first to-be-judged data point. Determine whether the first to-be-judged data point is located in the first tire area 6 in the corresponding wheel template, and if so, it is recorded as a second data point, and if not, it is recorded as a second to-be-judged data point. Determine whether the second to-be-judged data point is located in the second tire area 7 in the corresponding wheel template, and if so, it is recorded as a third data point, and if not, it is recorded as an invalid data point.

[0103] All the first data points form a first data point set, all the second data points form a second data point set, and all the third data points form a third data point set. According to the first data point set, the second data point set and the third data point set, an error minimization problem is established as follows:

[0104]

[0105] Wherein, X(k), Y(k) represent the coordinates of the data points in the first data point set, X(i), Y(i) represent the coordinates of the data points in the second data point set, X(j), Y(j) represent the coordinates of the data points in the third data point set, L1 represents the first data point set, L2 represents the second data point set, and L3 represents the third data point set. represents the coordinates of the first data point, represents the coordinates of the second data point, represents the coordinates of the third data point, and the values of A, B, C and D are obtained by solving the error minimization problem using a convex optimization method.

[0106] The values of A and D are used as parameters of the first straight line equation to obtain a first straight line, the values of A and B are used as parameters of the second straight line equation to obtain a second straight line, and the values of A and C are used as parameters of the third straight line equation to obtain a third straight line.

[0107] The intersection of the first straight line and the second straight line is taken as the left wheel position of the vehicle, the intersection of the first straight line and the third straight line is taken as the right wheel position of the vehicle, and the slope of the first straight line is taken as the wheel pose of the vehicle.

[0108] Specifically, the multi-straight line fitting is modeled as an error minimization problem, so as to obtain a first straight line fitted by laser data points in a tire face region, a second straight line fitted by laser data points in a first wheel region, and a third straight line fitted by laser data points in a second wheel region, an equation expression of the first straight line is Y+AX+D=0, an equation expression of the second straight line is Ay-x+B=0, and an equation expression of the third straight line is Ay-x+C=0, an intersection of the first straight line and the second straight line can be calculated according to the equation expressions of the first straight line and the second straight line, the intersection is a left wheel position of the vehicle, an intersection of the first straight line and the third straight line can be calculated according to the equation expressions of the first straight line and the third straight line, the intersection is a right wheel position of the vehicle, since the first straight line is fitted according to the first data points in the tire face region, a slope of the first straight line represents a wheel posture of the vehicle, and a pose of the vehicle is determined according to the left wheel position, the right wheel position, and the wheel posture, so as to facilitate the vehicle lifting robot to move to the vehicle according to the pose of the vehicle and enter below a chassis of the vehicle to perform a vehicle lifting operation.

[0109] Further, after the S6, the following steps are further included:

[0110] According to a preset third conversion formula, the first positioning coordinate, and a target wheel sample, the left wheel position is converted into a world coordinate system to obtain an actual left wheel position, wherein the target wheel sample is a wheel sample corresponding to a target vehicle laser point set;

[0111] According to a preset third conversion formula, the first positioning coordinate, and a target wheel sample, the right wheel position is converted into a world coordinate system to obtain an actual right wheel position;

[0112] The third conversion formula is,

[0113] wherein x represents a numerical value of the actual left wheel position or the actual right wheel position in an X dimension of the world coordinate system, y represents a numerical value of the actual left wheel position or the actual right wheel position in a Y dimension of the world coordinate system, xw2 represents a numerical value of the target wheel sample in an X dimension of a vehicle lifting robot coordinate system, yw2 represents a numerical value of the target wheel sample in a Y dimension of the vehicle lifting robot coordinate system, thetaw2 represents a numerical value of the target wheel sample in a theta dimension of the vehicle lifting robot coordinate system, xR represents a numerical value of the vehicle lifting robot in the X dimension of the world coordinate system, yR represents a numerical value of the vehicle lifting robot in the X dimension of the world coordinate system, thetaR represents a numerical value of the vehicle lifting robot in the theta dimension of the world coordinate system, xLFS represents a numerical value of the left wheel position or the right wheel position in an X dimension of a template coordinate system, and yLFS represents a numerical value of the left wheel position or the right wheel position in a Y dimension of the template coordinate system.

[0114] In practical applications, (x, y) represents the coordinates of the actual left wheel position or the actual right wheel position in the world coordinate system, (xw2, yw2, thetaw2) represents the coordinates of the target wheel sample in the robot coordinate system, (xR, yR, thetaR) represents the first positioning coordinates of the robot in the world coordinate system, (xLFS, yLFS) represents the coordinates of the left wheel position or the right wheel position in the template coordinate system, the left wheel position or the right wheel position is converted to the robot coordinate system according to the conversion relationship between the target wheel sample and the template coordinate system, and then the left wheel position or the right wheel position in the robot coordinate system is converted to the world coordinate system according to the conversion relationship between the robot coordinate system and the world coordinate system to obtain the actual left wheel position or the actual right wheel position, so that the robot is conveniently moved to the position of the vehicle to carry the vehicle.

[0115] The vehicle pose positioning method provided by the application first filters the laser point cloud data to obtain a first vehicle laser point set through the parking space data, obtains a vehicle positioning frame according to the first vehicle laser point set, greatly reduces the detection range, then segments the vehicle positioning frame to obtain a plurality of wheel samples through a preset wheel track range and preset segmentation data, then converts the laser data points in the first laser point set according to each wheel sample to obtain a corresponding second vehicle laser point set, and then compares each second vehicle laser point set with a corresponding wheel template, takes the second vehicle laser point set with the highest similarity as the target vehicle laser point set, takes the wheel template corresponding to the target vehicle laser point set as the target wheel template, and performs point cloud classification and error minimization fitting on the target vehicle laser point set according to the target wheel template to obtain the left wheel position, the right wheel position and the wheel pose of the vehicle, that is, the target vehicle pose is obtained, so that the robot can carry the vehicle, the vehicle is mainly positioned by using the wheel track and the wheel track center, the positioning is accurate and fast, and therefore the positioning speed of the vehicle is improved.

[0116] As shown in Figure 3 The application further provides a vehicle pose positioning system, which comprises:

[0117] A data acquisition module 10 is configured to acquire parking space data and laser point cloud data containing wheels of a vehicle, wherein the parking space data is parking space position data of a parking space where the vehicle is parked.

[0118] A data filtering module 20 is configured to filter the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtain a vehicle positioning frame according to the first vehicle laser point set.

[0119] A sample determination module 30 is configured to obtain a plurality of wheel samples according to the vehicle positioning frame, a preset wheel track range and preset segmentation data.

[0120] The coordinate conversion module 40 is configured to convert the first vehicle laser point set into a corresponding second vehicle laser point set in a template coordinate system according to a preset wheel template set and each wheel sample, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval.

[0121] The data comparison module 50 is configured to determine the corresponding second vehicle laser point set according to each wheel sample, determine the corresponding wheel template from the wheel template set, compare each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and select a target vehicle laser point set from all second vehicle laser point sets according to all comparison results.

[0122] The pose determination module 60 is configured to perform point cloud classification and error minimization fitting on the target vehicle laser point set according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose.

[0123] The specific definitions of the vehicle pose positioning system can refer to the definitions of the vehicle pose positioning method in the foregoing, and will not be described here. Each module of the vehicle pose positioning method system can be realized by software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0124] In one embodiment, a computer device, which can be a server, has an internal structure as shown in Figure 4 The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a vehicle pose positioning method.

[0125] Those skilled in the art can understand that Figure 4 the structure shown in the foregoing merely is a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the diagram, or combine certain components, or have a different arrangement of components.

[0126] In one embodiment, a computer device is provided, comprising a memory and a processor, the memory stores a computer program, and the processor implements a vehicle pose positioning method when executing the computer program.

[0127] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement a vehicle pose positioning method.

[0128] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, the processes of the above-mentioned embodiments can be included. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0129] Each technical feature of the above embodiments can be combined arbitrarily, and in order to make the description simple, not all possible combinations of each technical feature in the above embodiments are described, however, as long as the combination of these technical features does not exist, it should be considered as the scope of the present disclosure.

Claims

1. A vehicle pose positioning method, characterized by, Comprise: S1, acquiring parking space data and containing laser point cloud data of wheels in the vehicle, the parking space data being parking position data of a parking space where the vehicle is parked; S2, filtering the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtaining a vehicle positioning frame according to the first vehicle laser point set; S3, obtaining a plurality of wheel samples according to the vehicle positioning frame, a preset wheelbase range and preset segmentation data; S4, converting the first vehicle laser point set into a template coordinate system according to a preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval; S5, determining the corresponding second vehicle laser point set according to each wheel sample, and determining the corresponding wheel template from the wheel template set, comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results; S6, performing point cloud classification and error minimization fitting on the target vehicle laser point set according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose.

2. The vehicle pose localization method of claim 1, wherein, The parking space data is obtained from a parking lot coordinate system constructed by a SLAM method; the vehicle laser data points are generated based on a laser radar coordinate system; S2 comprises the following sub-steps: S21, calibrating the parking lot and the laser radar to determine the correspondence between the parking lot and the laser radar, and converting the parking space data into the laser radar coordinate system to obtain parking space laser data according to the correspondence; S22, comparing the parking space laser data with the laser point cloud data to filter out the laser point cloud data located in the parking space to obtain a plurality of laser data points to form a parking space laser point set; S23, reducing the parking space laser data according to a preset ratio to obtain reduced parking space laser data, filtering the parking space laser point set according to the reduced parking space laser data to obtain a plurality of filtered laser data points to form a filtered laser point set; S24, judging whether the difference between the number of laser data points of the parking space laser point set and the filtered laser point set is greater than a preset threshold, if yes, taking the parking space laser point set as the first vehicle laser point set, and executing sub-step S25, if not, taking the reduced parking space laser data as the parking space laser data, taking the filtered laser point set as the parking space laser point set, and returning to execute sub-step S23; S25, calculating the first vehicle laser point set by using a minimum bounding box algorithm to obtain a vehicle positioning frame.

3. The vehicle pose localization method of claim 2, wherein, The preset segmentation data comprises a first segmentation number in the x dimension of the laser radar coordinate system, a second segmentation number in the y dimension of the laser radar coordinate system, a third segmentation number in the theta dimension of the laser radar coordinate system, and a fourth segmentation number in the preset wheelbase range; S3 comprises the following sub-steps: S31, segmenting the vehicle positioning frame in the x dimension of the laser radar coordinate system according to the first segmentation number to obtain a plurality of first planes; S32, segmenting the vehicle positioning frame in the y dimension of the laser radar coordinate system according to the second segmentation number to obtain a plurality of second planes; S33, segmenting the vehicle positioning frame in the theta dimension of the laser radar coordinate system according to the third segmentation number to obtain a plurality of third planes; S34, obtaining a plurality of intersection points according to the intersection of the first plane, the second plane and the third plane; S35, selecting a plurality of sample wheelbases in a preset wheelbase range according to the fourth segmentation number; S36, combining each sample wheelbase with each intersection point to obtain a corresponding wheel sample.

4. The vehicle pose localization method of claim 3, wherein, The S4 comprises the following steps: S41, obtain a first positioning coordinate of the car-moving robot in a world coordinate system and a second positioning coordinate of the laser radar in a car-moving robot coordinate system, and convert a first vehicle laser point set to the car-moving robot coordinate system according to the first positioning coordinate, the second positioning coordinate, and a preset first conversion formula to obtain a first sub-vehicle laser point set, wherein the first conversion formula is, , Wherein, xm represents the value in the X dimension of the car moving robot coordinate system, ym represents the value in the Y dimension of the car moving robot coordinate system, xR represents the value of the car moving robot in the X dimension of the world coordinate system, yR represents the value of the car moving robot in the X dimension of the world coordinate system, thetaR represents the value of the car moving robot in the theta dimension of the world coordinate system, xL represents the value of the laser radar in the X dimension of the car moving robot coordinate system, yL represents the value of the laser radar in the Y dimension of the car moving robot coordinate system, thetaL represents the value of the laser radar in the theta dimension of the car moving robot coordinate system, xP represents the value in the X dimension of the laser radar coordinate system, and yP represents the value in the Y dimension of the laser radar coordinate system. S42, converting the first vehicle laser point set into a template coordinate system according to a preset second conversion formula and each of the wheel samples, the second conversion formula being, , Wherein, xw represents the value in the X dimension of the template coordinate system, yw represents the value in the Y dimension of the template coordinate system, xw1 represents the value of the intersection point in the wheel sample in the X dimension of the car moving robot coordinate system, yw1 represents the value of the intersection point in the wheel sample in the Y dimension of the car moving robot coordinate system, and thetaw1 represents the value of the intersection point in the wheel sample in the theta dimension of the car moving robot coordinate system.

5. The vehicle pose localization method of claim 4, wherein, The wheel template comprises a wheel area and a spacing area; the comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results, comprises: Comparing each second vehicle laser point set with the corresponding wheel template to determine whether each data laser point in the second vehicle laser point set is located in the wheel area of the corresponding wheel template, if yes, it is recorded as 1 point, if not, it is recorded as a to-be-compared data laser point, and it is determined whether the to-be-compared data laser point is located in the spacing area of the corresponding wheel template, if yes, it is recorded as-1 point, if not, it is recorded as 0 point; Adding the scores of all data laser points in each second vehicle laser point set to obtain the corresponding comparison result; Comparing the comparison results of all second vehicle laser point sets to select the second vehicle laser point set with the highest score as the target vehicle laser point set.

6. The vehicle pose localization method of claim 1, wherein, The wheel template is divided into a tire surface area, a first tire area and a second tire area; the S6 comprises the following sub-steps: The target vehicle laser point set is compared with the corresponding wheel template to determine whether each data laser point in the target vehicle laser point set is located in a tire surface area in the corresponding wheel template, and if so, the data laser point is recorded as a first data point, and if not, the data laser point is recorded as a first to-be-judged data point. It is determined whether the first to-be-judged data point is located in a first tire area in the corresponding wheel template, and if so, the first to-be-judged data point is recorded as a second data point, and if not, the first to-be-judged data point is recorded as a second to-be-judged data point. It is determined whether the second to-be-judged data point is located in a second tire area in the corresponding wheel template, and if so, the second to-be-judged data point is recorded as a third data point, and if not, the second to-be-judged data point is recorded as an invalid data point; All the first data points form a first data point set, all the second data points form a second data point set, and all the third data points form a third data point set. An error minimization problem is established according to the first data point set, the second data point set, and the third data point set as follows: ， Wherein, X(k), Y(k) represent the coordinates of the data points in the first data point set, X(i), Y(i) represent the coordinates of the data points in the second data point set, X(j), Y(j) represent the coordinates of the data points in the third data point set, L1 represents the first data point set, L2 represents the second data point set, and L3 represents the third data point set. The values of A, B, C, and D are obtained by solving the error minimization problem using a convex optimization method. The values of A and D are used as parameters of a first straight line equation to obtain a first straight line, the values of A and B are used as parameters of a second straight line equation to obtain a second straight line, and the values of A and C are used as parameters of a third straight line equation to obtain a third straight line. The intersection of the first straight line and the second straight line is used as the left wheel position of the vehicle, the intersection of the first straight line and the third straight line is used as the right wheel position of the vehicle, and the slope of the first straight line is used as the wheel pose of the vehicle.

7. The vehicle pose localization method of claim 6, wherein, The following steps are further included after the S6: Obtaining a first positioning coordinate of the vehicle moving robot in a world coordinate system; Converting the left wheel position to the world coordinate system to obtain an actual left wheel position according to a preset third conversion formula, the first positioning coordinate, and a target wheel sample, wherein the target wheel sample is a wheel sample corresponding to the target vehicle laser point set; Converting the right wheel position to the world coordinate system to obtain an actual right wheel position according to a preset third conversion formula, the first positioning coordinate, and a target wheel sample; Converting the right wheel position to the world coordinate system to obtain an actual right wheel position according to a preset third conversion formula, the first positioning coordinate, and a target wheel sample; The third conversion formula is, , Wherein, x represents the actual left wheel position or the actual right wheel position in the X dimension of the world coordinate system, y represents the actual left wheel position or the actual right wheel position in the Y dimension of the world coordinate system, xw2 represents the target wheel sample in the X dimension of the car robot coordinate system, yw2 represents the target wheel sample in the Y dimension of the car robot coordinate system, thetaw2 represents the target wheel sample in the theta dimension of the car robot coordinate system, xR represents the car robot in the X dimension of the world coordinate system, yR represents the car robot in the X dimension of the world coordinate system, thetaR represents the car robot in the theta dimension of the world coordinate system, xLFS represents the left wheel position or the right wheel position in the X dimension of the template coordinate system, and yLFS represents the left wheel position or the right wheel position in the Y dimension of the template coordinate system.

8. A vehicle pose positioning system, characterized by, Comprise: A data acquisition module for acquiring parking space data and laser point cloud data containing vehicle wheels, wherein the parking space data is the parking position data of the parking space where the vehicle is parked; A data filtering module for filtering the laser point cloud data according to the parking space data to obtain a first vehicle laser point set, and obtaining a vehicle positioning frame according to the first vehicle laser point set; A sample determination module for obtaining a plurality of wheel samples according to the vehicle positioning frame, a preset wheelbase range, and a preset segmentation data; A coordinate conversion module for converting the first vehicle laser point set to a template coordinate system according to a preset wheel template set and each wheel sample to obtain a corresponding second vehicle laser point set, wherein the wheel template set is constructed according to a preset wheelbase range and a preset interval; A data comparison module for determining the corresponding second vehicle laser point set according to each wheel sample, and determining the corresponding wheel template from the wheel template set, comparing each second vehicle laser point set with the corresponding wheel template to obtain a comparison result, and selecting a target vehicle laser point set from all second vehicle laser point sets according to all comparison results; A pose determination module for performing point cloud classification and error minimization fitting on the target vehicle laser point set according to the corresponding wheel template of the target vehicle laser point set to obtain a target vehicle pose. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8. The processor executes the computer program to realize the steps of the method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method of any one of claims 1 to 7.

Images