Search

CN-122009153-A - Parking track optimization method and vehicle

CN122009153ACN 122009153 ACN122009153 ACN 122009153ACN-122009153-A

Abstract

The invention discloses a parking track optimization method and a vehicle, wherein a short-section original track with a higher curvature at a starting end is deleted and replaced by a derivative track section with a smaller curvature and a longer length, so that the geometric characteristics of the track starting section are changed, the vehicle can start to move from a state with a smaller steering wheel angle, smooth starting of a steering wheel is realized, and parking experience and efficiency are improved. By carrying out translation and rotation alignment based on the original starting point state on the generated initial sub-track, the optimized track is ensured to start from the actual starting position and posture of the vehicle, the accuracy of the parking starting point is ensured, and execution errors caused by track deviation are avoided. The processed derivative sub-track and the original sub-track are fused, the generated first fusion sub-track is attached to the initial part and has the smooth characteristic after optimization, the end part returns to the accurate end point of the original track, and the positioning precision of the parking end point is maintained while the whole smooth and continuous process is ensured.

Inventors

  • Cao Jiateng
  • TU NINGNING
  • HOU MINGYANG
  • KANG SHUTAO

Assignees

  • 岚图汽车科技股份有限公司

Dates

Publication Date
20260512
Application Date
20260202

Claims (10)

  1. 1. A method of optimizing a parking trajectory, the method comprising: acquiring an original parking track corresponding to a target vehicle at a current position, wherein the original parking track comprises at least one section of original sub-track, and the original sub-track refers to a track section between positions of the target vehicle when the target vehicle is shifted for any two adjacent times; Executing a first optimization step on a starting end of a first section of original sub-track of the original parking track, where the current position is located, to obtain a first derivative track section, wherein the first section of original sub-track is a target original sub-track in the first optimization step, and a starting end of the first section of original sub-track is a first end in the first optimization step; combining the first derivative track section with the track section reserved in the first original sub-track section to obtain a first initial sub-track, translating and rotating the first initial sub-track based on track parameters of track feature points of the initial end of the first original sub-track before the track section is deleted, enabling a starting point of the first initial sub-track to coincide with the track feature points of the initial end of the first original sub-track before the track section is deleted to obtain a first derivative sub-track, and fusing the first original sub-track and the first derivative sub-track to obtain a first fused sub-track; Determining the first fusion sub-track as the target parking track after optimizing the target vehicle under the condition that the original parking track only comprises the first section of original sub-track; The first optimization step comprises deleting an original track section with a first preset length, which is located at a first end of a target original sub-track, and adding a derivative track section with a second preset length to the first end of the target original sub-track, wherein the first preset length is smaller than the second preset length, and the curvature of the derivative track section is smaller than that of the original track section.
  2. 2. The method for optimizing a parking trajectory according to claim 1, wherein deleting an original trajectory segment having a first predetermined length and where a first end of a target original sub-trajectory is located, and adding a derivative trajectory segment having a second predetermined length to the first end of the target original sub-trajectory, comprises: Dividing the original target sub-track into N-1 track units with preset sub-lengths, wherein the endpoints at two ends of each track unit are marked as track characteristic points, and the original target sub-track comprises N track characteristic points; starting from the track characteristic point of the target original sub-track at the first end, deleting an original track section with a first preset length, wherein the original track section is a track section formed by all track units from the first track characteristic point to the M+1th track characteristic point, M is a positive integer, and 2< M < N; Determining K derived feature points on one side, away from the M+1th track unit, of the M+1th track feature point based on track parameters corresponding to the M+1th track feature point, wherein a distance with a preset sub-length is reserved between every two adjacent feature points in the M+1th track feature point to the K derived feature point, the sum of straight line distances between every two adjacent feature points in the M+1th track feature point to the K derived feature point is a second preset length, the difference value of steering wheel angles corresponding to any two adjacent feature points in the K-1th derived feature point is the same, the steering wheel angles of all feature points in the K-1th derived feature point to the K derived feature point are 0, the vehicle heading angles of all feature points in the K-2th derived feature point to the K derived feature point are the same, K is a positive integer, K is more than 4, and M < K < N; And obtaining a derivative track segment based on the M+1th track feature point to the K derivative feature point.
  3. 3. The method for optimizing a parking trajectory of claim 2, wherein the trajectory parameters include trajectory coordinates, a vehicle heading angle, a steering wheel angle, and a driving direction; Based on the track parameter corresponding to the M+1th track feature point, determining K derivative feature points on one side of the M+1th track feature point away from the M+1th track unit, wherein the K derivative feature points comprise: The steering wheel angle of the M+1th track feature point is determined to be the steering wheel angle of a first derivative feature point, the difference value of the steering wheel angles corresponding to any two adjacent feature points from the first derivative feature point to the K-1th derivative feature point is the same, the steering wheel angles of the K-1th derivative feature point and the K-th derivative feature point are 0, and the vehicle course angles of all feature points from the K-2th derivative feature point to the K-th derivative feature point are the same; i traverses from 1 to K-1 in sequence, a first step is performed for i, the first step comprising: Determining an ith driving radius of the target vehicle at the ith derivative feature point based on the steering wheel angle of the ith derivative feature point and the wheelbase of the target vehicle; determining an ith difference value of a vehicle course angle between an ith derivative feature point and an ith-1 derivative feature point based on a preset sub-length between the ith derivative feature point and the ith-1 derivative feature point and the ith running radius; Determining the vehicle course angle of the ith derivative feature point based on the vehicle course angle of the ith-1 derivative feature point and the ith difference value; determining the track coordinates of the ith derivative feature point based on the ith running radius, the track coordinates of the ith-1 derivative feature point and the ith difference value; After i taking K-1 and executing the first step, determining the track coordinates of the Kth derivative feature point based on the track coordinates of the Kth derivative feature point and the heading angle of the vehicle and the preset sub-length between the Kth derivative feature point and the Kth derivative feature point.
  4. 4. The method for optimizing a parking trajectory according to claim 1, wherein fusing the first segment of original sub-trajectory and the first derivative sub-trajectory to obtain a first fused sub-trajectory comprises: Dividing the first section of original sub-track into N-1 first track units with preset sub-lengths, wherein the end points at the two ends of each first track unit are marked as first track characteristic points, each first track characteristic point comprises a first accumulated length value, the first accumulated length value refers to the sum of the lengths of all first track units before the first track characteristic point, and the first section of original sub-track comprises N first track characteristic points; dividing the derivative total length of the first derivative sub-track into N-1 second track units, wherein the end points at the two ends of each second track unit are marked as second track characteristic points, each second track characteristic point comprises a second accumulated length value, the second accumulated length value refers to the sum of the lengths of all second track units before the second track characteristic point, and the first derivative sub-track comprises N second track characteristic points; Determining a first deviation degree of each second track characteristic point relative to the first track characteristic points by taking each first track characteristic point as a reference, and carrying out interpolation processing on each second track characteristic point based on the first deviation degree to obtain N first target interpolation points; Determining a second deviation degree of each first track characteristic point relative to each second track characteristic point by taking each second track characteristic point as a reference, and carrying out interpolation processing on each first track characteristic point based on the second deviation degree to obtain N second target interpolation points; And obtaining N fusion characteristic points based on the N first target interpolation points and the N second target interpolation points, and determining a first fusion sub-track based on the N fusion characteristic points.
  5. 5. The method for optimizing a parking trajectory of claim 4, wherein determining a first degree of deviation of each second trajectory feature point with respect to each first trajectory feature point based on each first trajectory feature point, and interpolating each second trajectory feature point based on the first degree of deviation to obtain N first target interpolation points, comprises: Sequentially taking each first track characteristic point in the first section of original sub-track as a first target characteristic point and executing a second step to obtain N first target interpolation points corresponding to the first derivative sub-track; wherein the second step comprises: Screening out a second target characteristic point with the minimum second accumulated length value from the characteristic points with the second accumulated length value larger than the first accumulated length of the first target characteristic point in the first derivative sub-track; screening a third target characteristic point with the maximum second accumulated length value from the characteristic points with the second accumulated length value smaller than the first accumulated length of the first target characteristic point in the first derivative sub-track; Determining a first degree of deviation between a second target feature point and a third target feature point based on a first accumulated length value of the first target feature point, a second accumulated length value of the second target feature point, and a second accumulated length value of the third target feature point; And determining a first target interpolation point corresponding to the second target feature point and the third target feature point based on the track parameter of the second target feature point, the track parameter of the third target feature point and the first deviation degree.
  6. 6. The method for optimizing a parking trajectory of claim 4, wherein determining a second degree of deviation of each first trajectory feature point with respect to each second trajectory feature point based on each second trajectory feature point, and interpolating each first trajectory feature point based on the second degree of deviation to obtain N second target interpolation points, comprises: Sequentially taking each second track characteristic point in the first derivative sub-track as a fourth target characteristic point and executing a third step to obtain N second target interpolation points corresponding to the first section original sub-track; Wherein the third step comprises: screening a fifth target characteristic point with the minimum first accumulated length value from the characteristic points with the first accumulated length value larger than the second accumulated length of the fourth target characteristic point in the first section of original sub-track; Screening a sixth target characteristic point with the maximum first accumulated length value from the characteristic points with the first accumulated length value smaller than the second accumulated length of the fourth target characteristic point in the first section of original sub-track; Determining a second degree of deviation between a fifth target feature point and a sixth target feature point based on a second cumulative length value of the fourth target feature point, a first cumulative length value of the fifth target feature point, and a first cumulative length value of the sixth target feature point; And determining a second target interpolation point corresponding to the fifth target feature point and the sixth target feature point based on the track parameter of the fifth target feature point, the track parameter of the sixth target feature point and the second deviation degree.
  7. 7. The method for optimizing a parking trajectory of claim 4, wherein obtaining N fused feature points based on the N first target interpolation points and the N second target interpolation points, and determining a first fused sub-trajectory based on the N fused feature points, comprises: taking the first a first target interpolation points in the N first target interpolation points as the first a fusion characteristic points of the first fusion sub-track, taking the last b second target interpolation points in the N second target interpolation points as the last b fusion characteristic points of the first fusion sub-track, wherein a and b are positive integers, a is less than or equal to 8, and b is less than or equal to 8; For the (a+1) th first target interpolation point to the (N-b) th first target interpolation point in the N first target interpolation points and the (a+1) th second target interpolation point to the (N-b) th second target interpolation point in the N second target interpolation points, determining a first fusion coefficient between the first target interpolation point and the second target interpolation point which are corresponding one to one each based on a first formula, wherein the first formula is as follows: coef=0.5*{1-cos[(i-a)*PI/(N-a-b)]}; The coef is a first fusion coefficient of the i-th corresponding first target interpolation point and the second target interpolation point in the N first target interpolation points and the N second target interpolation points, cos [ (i-a) PI/(N-a-b) ] represents a cosine function of (i-a) PI/(N-a-b), i represents a serial number of the i-th interpolation point in the extending direction from the starting point to the end point along the first segment original sub-track in the N first target interpolation points and the N second target interpolation points, i is a positive integer, and a < i < N-b+1; PI represents PI; determining an a+1th fusion characteristic point to an N-th fusion characteristic point in a first fusion sub-track based on a+1th first target interpolation point to an N-b first target interpolation point in the N first target interpolation points, a+1th second target interpolation point to an N-b second target interpolation point in the N second target interpolation points and first fusion coefficients corresponding to each pair of the first target interpolation points and the second target interpolation points; And determining a first fusion sub-track based on the first a fusion characteristic points, the (a+1) th fusion characteristic points, the (N-b) th fusion characteristic points and the last b fusion characteristic points.
  8. 8. The method for optimizing a parking trajectory according to claim 1, wherein after fusing the first original sub-trajectory and the first derivative sub-trajectory to obtain a first fused sub-trajectory, in a case where the original parking trajectory includes the first original sub-trajectory and at least one second original sub-trajectory, the method further includes: Aiming at a pair of sub-tracks with intersection points in each two sections of the first fused sub-track and at least one section of second original sub-track after the first section of original sub-track is updated, the pair of sub-tracks comprises a front sub-track and a rear sub-track, and a second optimization step is carried out on the pair of sub-tracks corresponding to each intersection point one by one according to the sequence of the original parking track, so that fused sub-tracks corresponding to the first fused sub-track and the at least one section of second original sub-track are obtained; Determining a parking track formed by each fusion sub-track as a target parking track after optimizing the target vehicle; the second optimizing step includes: performing a first optimization step on the ending end of the previous sub-track to obtain a previous derived track segment, and combining the previous derived track segment with the track segment reserved in the previous sub-track to obtain a previous initial sub-track, wherein the previous sub-track is the target original sub-track in the first optimization step, and the ending end in the previous sub-track is the first end in the first optimization step; performing a first optimization step on the initial end of the subsequent sub-track to obtain a subsequent derived track section, and combining the subsequent derived track section with a track section reserved in the subsequent sub-track to obtain a subsequent initial sub-track, wherein the subsequent sub-track is a target original sub-track in the first optimization step, and the initial end in the subsequent sub-track is a first end in the first optimization step; Determining an intermediate point based on a preceding end point of the ending end of the preceding initial sub-track and a following end point of the starting end of the following initial sub-track, the intermediate point being on a line between the preceding end point and the following end point; translating and rotating the prior initial sub-track based on the intermediate point so that the front end point of the end of the prior initial sub-track coincides with the intermediate point to obtain a prior derivative sub-track; Translating and rotating the subsequent initial sub-track based on the intermediate point, so that the subsequent end point of the initial end of the subsequent initial sub-track coincides with the intermediate point to obtain a subsequent derived sub-track, and fusing the subsequent derived sub-track and the subsequent sub-track to obtain a subsequent fused sub-track; And if the post-fusion sub-track corresponds to a second original sub-track which is not the last section of the original parking track, taking the post-fusion sub-track as a previous sub-track for executing a second optimization step next time.
  9. 9. The parking trajectory optimization method according to claim 8, wherein determining the intermediate point based on the preceding end point of the ending end of the preceding initial sub-trajectory and the following end point of the starting end of the following initial sub-trajectory includes: determining the sum of the preceding track length of the preceding initial sub-track and the following track length of the following initial sub-track as the total sub-track length; Determining the ratio of the length of the previous track to the total length of the sub-track as a second fusion coefficient; Determining the product of the second fusion coefficient and the track parameter at the rear end point as first data; Determining the difference value between the 1 and the second fusion coefficient as a third fusion coefficient; determining the product of the third fusion coefficient and the track parameter at the front end point as second data; the sum of the first data and the second data is determined as a trajectory parameter of the intermediate point.
  10. 10. A vehicle, characterized by comprising: A processor; a memory for storing the processor-executable instructions; wherein the processor is configured to execute to implement a parking trajectory optimization method as claimed in any one of claims 1 to 9.

Description

Parking track optimization method and vehicle Technical Field The invention relates to the technical field of automobiles, in particular to a parking track optimization method and a vehicle. Background During automatic parking, the vehicle is typically driven on a planned trajectory. However, a common parking trajectory often requires a large steering wheel angle at the starting point, resulting in that the vehicle must first turn the steering wheel in place to a certain angle before starting to move, which is not consistent with human driving habits and takes a long time. How to improve the smoothness and consistency of the vehicle in the parking start is a problem which needs to be solved currently. Disclosure of Invention According to the method and the vehicle for optimizing the parking track, the technical problems that in the prior art, the curvature of the parking track at the starting point is not zero, and the vehicle can start or shift gears only by driving the steering wheel in situ, so that parking experience is poor and efficiency is low are solved, a new track with the curvature smoothly transited to zero at the starting point and the shifting point is generated on the premise that the overall shape and the end point of the parking track are not changed, the vehicle can be turned while moving (driving the steering wheel), dynamic smooth gear shifting is realized, and the operation naturalness, comfort and overall efficiency of parking are effectively improved. In a first aspect, the present application provides a method for optimizing a parking trajectory, the method comprising: acquiring an original parking track corresponding to a target vehicle at a current position, wherein the original parking track comprises at least one section of original sub-track, and the original sub-track refers to a track section between positions of the target vehicle when the target vehicle is shifted for any two adjacent times; Executing a first optimization step on a starting end of a first section of original sub-track of the original parking track, where the current position is located, to obtain a first derivative track section, wherein the first section of original sub-track is a target original sub-track in the first optimization step, and a starting end of the first section of original sub-track is a first end in the first optimization step; combining the first derivative track section with the track section reserved in the first original sub-track section to obtain a first initial sub-track, translating and rotating the first initial sub-track based on track parameters of track feature points of the initial end of the first original sub-track before the track section is deleted, enabling a starting point of the first initial sub-track to coincide with the track feature points of the initial end of the first original sub-track before the track section is deleted to obtain a first derivative sub-track, and fusing the first original sub-track and the first derivative sub-track to obtain a first fused sub-track; Determining the first fusion sub-track as the target parking track after optimizing the target vehicle under the condition that the original parking track only comprises the first section of original sub-track; The first optimization step comprises deleting an original track section with a first preset length, which is located at a first end of a target original sub-track, and adding a derivative track section with a second preset length to the first end of the target original sub-track, wherein the first preset length is smaller than the second preset length, and the curvature of the derivative track section is smaller than that of the original track section. In a second aspect, the present application provides a vehicle comprising: A processor; a memory for storing the processor-executable instructions; wherein the processor is configured to execute to implement a method of optimizing a parking trajectory as provided in the first aspect. One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages: According to the parking track optimization method provided by the embodiment of the application, the obvious technical effect is realized by acquiring the original parking track comprising at least one section of original sub-track and executing a specific first optimization step on the initial end of the first section of original sub-track. Firstly, by deleting a short-section original track with a higher curvature at the starting end and replacing the short-section original track with a derivative track section with a smaller curvature and a longer length, the geometric characteristic of the track starting section is fundamentally changed, so that a vehicle can start moving from a state with a smaller steering wheel angle, the technical bottleneck that the vehicle can start only by steering wheel in situ in traditio