Path planning method, device and vehicle

By obtaining the reference path curve and optimizing according to the preset target path function, combining obstacle avoidance boundary conditions and curvature constraints, the problem of path curvature exceeding the vehicle limit is solved, and reasonable path planning is achieved to ensure the executability of the path.

CN113483776BActive Publication Date: 2024-11-26GUANGZHOU XIAOPENG CONNECTIVITY TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110747895.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-07-01
Publication Date
2024-11-26
Estimated Expiration
2041-07-01

AI Technical Summary

Technical Problem

In the path planning of automobile autonomous driving, the prior art fails to effectively constrain the path curvature, resulting in the vehicle limit exceeding and a reasonable and executable driving path cannot be given.

Method used

By obtaining the reference path curve, collecting the reference path points, and optimizing according to the preset target path function, combining obstacle avoidance boundary conditions and curvature constraints, the location information of the optimized path points is obtained, thereby generating a reasonable planned path curve to avoid curvature exceeding the curvature Vehicle limit.

Benefits of technology

A reasonable and executable path planning is achieved, which avoids the problem of path curvature exceeding the vehicle limit, and improves the feasibility and accuracy of path planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113483776B_ABST
    Figure CN113483776B_ABST
Patent Text Reader

Abstract

The present application is about a path planning method, device and automobile. The method includes: obtaining a reference path curve; collecting reference path points of the reference path curve; obtaining the position information of the optimized path points according to the position information of the reference path points and the preset target path function; wherein the preset target path function is constrained according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions; obtaining the planned path curve according to the position information of the optimized path points. The solution provided by the present application can avoid the curvature of the planned path curve from exceeding the vehicle limit and provide a reasonable path that the vehicle can drive.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a path planning method, device and automobile. Background Art

[0002] In the path planning of the car's autonomous driving, in order to clarify the position of the car relative to the road, the autonomous driving adopts the Frenet coordinate system for path planning. In the Frenet coordinate system, the center line of the road is used as the reference line, and the tangent vector and normal vector of the reference line are used to establish the coordinate system. Therefore, based on the position of the reference line, the longitudinal distance coordinate along the tangent direction of the reference line and the lateral distance coordinate along the normal direction of the reference line that deviates from the reference line can be used to describe the path planning.

[0003] In related technologies, after a safe and collision-free path curve is established, the path curve is optimized through algorithms such as piecewise polynomials and piecewise Bezier curves. However, this optimization process has complex mathematical derivations and a large amount of calculations, and does not constrain the curvature of the path curve. As a result, when the curvature of the reference line exceeds the vehicle limit, it is impossible to provide a reasonable path that the vehicle can drive. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related art, the present application provides a path planning method, device and automobile. The path planning method can avoid the curvature of the planned path curve exceeding the vehicle limit and provide a reasonable path that the vehicle can drive.

[0005] The present application provides a path planning method in the first aspect, which comprises:

[0006] Obtain a reference path curve;

[0007] Collecting reference path points of the reference path curve;

[0008] According to the position information of the reference path point and the preset target path function, the position information of the optimized path point is obtained; wherein the preset target path function is constrained according to the preset obstacle avoidance boundary condition and the preset curvature constraint condition;

[0009] A planned path curve is obtained according to the position information of the optimized path points.

[0010] In one embodiment, obtaining a reference path curve includes:

[0011] In the Cartesian coordinate system, the center line of the road is taken as the road reference curve;

[0012] The reference path curve is obtained in a Frenet coordinate system according to the road reference curve.

[0013] In one embodiment, the collecting of reference path points of the reference path curve includes:

[0014] In the Frenet coordinate system, the longitudinal distance along the reference path curve is evenly divided to obtain the longitudinal distance coordinates of the reference path points.

[0015] In one embodiment, obtaining the position information of the optimized path point according to the position information of the reference path point and a preset target path function includes:

[0016] Calculate and obtain the lateral distance coordinate of the optimized path point according to the preset target path function and the longitudinal distance coordinate of the reference path point;

[0017] The position information of the optimized path point is obtained according to the lateral distance coordinate and the longitudinal distance coordinate of the reference path point.

[0018] In one embodiment, the preset target path function performs cost evaluation based on at least one of the following items: the minimum cost of the distance between the planned path curve and the reference path curve, the cost of the planned path curve being close to the road reference curve, the second-order smoothness cost of the planned path curve, and the third-order smoothness cost of the planned path curve.

[0019] In one embodiment, the obstacle avoidance boundary condition includes at least one of the following constraints:

[0020] Vehicle drivable boundary constraints, starting path point boundary constraints, ending path point boundary constraints and path point boundary constraints in the Frenet coordinate system.

[0021] In one embodiment, the preset curvature constraint condition is obtained according to a curvature function of a matching point of the road reference curve; wherein the matching point is obtained according to a corresponding mapping of the reference path point on the road reference curve.

[0022] A second aspect of the present application provides a path planning device, comprising:

[0023] A reference curve acquisition module, used to acquire a reference path curve;

[0024] A reference path point acquisition module, used for acquiring reference path points of the reference path curve;

[0025] An optimized path point acquisition module, used to obtain the position information of the optimized path point according to the position information of the reference path point and a preset target path function; wherein the preset target path function is constrained according to a preset obstacle avoidance boundary condition and a preset curvature constraint condition;

[0026] The planned path acquisition module is used to obtain the planned path curve according to the position information of the optimized path points.

[0027] A third aspect of the present application provides a car, which includes the path planning device described in any of the above embodiments.

[0028] A fourth aspect of the present application provides a non-transitory machine-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0029] The technical solution provided by this application may have the following beneficial effects:

[0030] The path planning method of the present application obtains a reference path curve, collects reference path points of the reference path curve, and obtains the position information of the optimized path points based on the position information of the reference path points and the preset target path function, wherein the preset target path function is constrained according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions, and the planned path curve is obtained according to the position information of the optimized path points. Such a design can obtain the position information of the optimized path points by constraining the preset curvature constraint conditions through the preset target path function, and then obtain the planned path curve, so as to avoid the curvature of the planned path curve exceeding the vehicle limit, thereby providing a reasonable path that the vehicle can drive.

[0031] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0033] Figure 1 is a flow chart of a path planning method shown in an embodiment of the present application;

[0034] Figure 2 is a flow chart of a path planning method according to another embodiment of the present application;

[0035] Figure 3 is a schematic diagram of a Cartesian coordinate system shown in an embodiment of the present application;

[0036] Figure 4 is a schematic diagram of the Frenet coordinate system shown in the embodiment of the present application;

[0037] Figure 5is a planned path curve and a partially enlarged schematic diagram of the path planning method shown in an embodiment of the present application;

[0038] Figure 6 is a schematic diagram comparing the direction angles before and after optimization of the path planning method shown in the embodiment of the present application;

[0039] Figure 7 is a schematic diagram of curvature comparison before and after optimization of the path planning method shown in the embodiment of the present application;

[0040] Figure 8 It is a schematic diagram of the structure of the path planning device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0042] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0043] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0044] In the path planning of autonomous driving, in order to clarify the position of the car relative to the road, autonomous driving uses the Frenet coordinate system for path planning. In related technologies, after establishing a safe and collision-free path curve, the path curve is optimized through algorithms such as piecewise polynomials and piecewise Bezier curves. However, this optimization process has complex mathematical derivations and a large amount of calculations, and does not constrain the curvature of the path curve, resulting in the curvature of the reference line exceeding the vehicle limit, and it is impossible to give a reasonable path that the vehicle can drive.

[0045] In response to the above problems, an embodiment of the present application provides a path planning method, which can prevent the curvature of the planned path curve from exceeding the vehicle limit and provide a reasonable path that the vehicle can drive.

[0046] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0047] See also Figure 1 The path planning method of this embodiment includes:

[0048] Step S110, obtaining a reference path curve.

[0049] The automatic driving process of a car is the process in which the car automatically drives from the preset starting point of the road to the preset end point. Between the preset starting point and the preset end point, countless travel paths can be formed. It can be understood that the corresponding reference path curve is set based on the requirements of safe obstacle avoidance and navigation routes. The reference path curve is a rough path curve to distinguish it from the planned path curve finally obtained for actual execution by the car. By using the reference path curve as a reference, the reference path curve is used as a reference and adjusted, so as to facilitate the subsequent steps to obtain a planned path curve that meets the requirements.

[0050] Step S120, collecting reference path points of the reference path curve.

[0051] Among them, the reference path point is the point that the reference path curve passes through. When collecting, it can be obtained in sequence along the extension direction of the reference path curve. The number of reference path points can be set according to the length of the reference path curve. For example, the longer the length of the reference path curve is, the more the number of reference path points is. Further, the interval distance between two adjacent reference path points on the reference path curve can be set according to the requirement of the adjustment accuracy of the planned path curve. The smaller the distance between two adjacent reference path points on the reference path curve, the higher the adjustment accuracy of the planned path curve. It can be understood that the curvature of the curve segment between two adjacent reference path points can be calculated according to the relevant technology. According to the reference path points, the corresponding optimized path points in the subsequent steps are obtained, so as to facilitate the adjustment of the curvature by adjusting the optimized path points. In other words, the more frequently the reference path points are collected, the more reference path points are, the more corresponding optimized path points are obtained in the subsequent steps, and the more corresponding curve segments with adjustable curvature are, which helps to achieve the curvature optimization of the planned path curve.

[0052] Step S130, obtaining the position information of the optimized path point according to the position information of the reference path point and the preset target path function, wherein the preset target path function is constrained according to the preset obstacle avoidance boundary condition and the preset curvature constraint condition.

[0053] In the related art, the position information of the reference path point in the coordinate system where the reference path curve is located can be obtained according to the position of the reference path point on the reference path curve. According to the position information of the reference path point and the preset target path function, the position information of the optimized path point in the coordinate system can be obtained.

[0054] Among them, the preset target path function is constrained according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions, so that the position information of the optimized path points is constrained accordingly, so that the obtained optimized path points can meet the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions, so that the planned path curve obtained in the subsequent steps meets the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions. For example, the preset obstacle avoidance boundary conditions can constrain the position information of the optimized path points according to the position information of the obstacle to avoid the optimized path points being set at the position of the obstacle. The preset curvature constraint condition can indirectly constrain the curvature of the planned path curve formed by the optimized path points by constraining the position information of the optimized path points. The preset curvature constraint condition can be set according to the steering limit of the car, and / or can be set according to the comfort level of the passengers to avoid excessive curvature causing discomfort to the user.

[0055] Furthermore, the target path function may be a function that measures the cost of obtaining the optimized path point, or a function that measures the degree of fit between the optimized path point and the reference path point, or a function that measures the smoothness of the curve formed by the optimized path point. The preset target function can reach an optimal value based on the location information of the optimized path point, and the optimal value may be the maximum value or minimum value of the preset target function. The location information of the optimized path point is determined based on the optimal value of the preset target function.

[0056] Step S140, obtaining a planned path curve according to the position information of the optimized path points.

[0057] By fitting the optimized path points, a planned path curve with the optimized path points is generated. The optimized path points are configured to avoid obstacles through the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions of step S130, and the curvature of the planned path curve can be prevented from exceeding the limit of the vehicle, making the planned path curve more reasonable and feasible.

[0058] In summary, the path planning method of the present application obtains the reference path curve, collects the reference path points of the reference path curve to obtain the position information of the reference path points, obtains the position information of the optimized path points according to the position information of the reference path points and the preset target path function, wherein the preset target path function is constrained according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions, and obtains the planned path curve according to the position information of the optimized path points. Such a design can obtain the position information of the optimized path points by constraining the preset curvature constraint conditions through the preset target path function, and then obtain the planned path curve, so as to avoid the curvature of the planned path curve exceeding the vehicle limit, and provide a reasonable path that the vehicle can drive.

[0059] To further understand the path planning method of this application, see Figure 2 The path planning method of this embodiment includes:

[0060] Step S210: In the Cartesian coordinate system, the center line of the road is used as a road reference curve; and according to the road reference curve, a reference path curve is obtained in the Frenet coordinate system.

[0061] It is understood that before the car can perform automatic driving, it is necessary to determine the road it is currently traveling on. In order to ensure driving safety and facilitate identification and positioning, the center line of the road it is currently traveling on is used as the road reference curve. It is understood that the center line of the road is parallel to the extension direction of the road, that is, the curvature of the center line of the road is the same as that of the road. Using the center line of the road as the road reference curve can accurately serve as the curvature reference benchmark for the planned path curve.

[0062] In the related art, the road reference curve is generally described in a Cartesian coordinate system. Figure 3 , in the Cartesian coordinate system, according to the road reference curve a, the reference path curve b in the Cartesian coordinate system can be obtained. In order to more intuitively represent the position of the car on the current road, according to the relative position information of the road reference curve a and the reference path curve b in the Cartesian coordinate system, the corresponding Frenet coordinate system is established through conversion. See also Figure 3 and Figure 4 ,according to Figure 3 The relative position information of the reference path curve b in the tangent direction and normal direction of the road reference curve a is Figure 4 The corresponding reference path curve b is generated in the Frenet coordinate system. Therefore, the relevant description and constraints in the subsequent steps are performed based on the road reference curve a in the Cartesian coordinate system and the reference path curve b in the Frenet coordinate system.

[0063] Step S220 , in the Frenet coordinate system, evenly divide the longitudinal distance along the reference path curve, collect reference path points of the reference path curve, and obtain longitudinal distance coordinates corresponding to the reference path points.

[0064] It can be understood that in order to obtain multiple reference path points, the reference path curve with a preset longitudinal distance can be evenly divided according to the preset longitudinal distance difference. It can be understood that based on the fact that the length of the road is known in advance, the preset longitudinal distance of the reference path curve is known in advance, and the preset longitudinal distance of the reference path curve is the total length of the reference path curve along the tangent direction of the road reference curve. In order to achieve uniform division, a preset longitudinal distance difference can be set. According to the preset longitudinal distance difference, the longitudinal distance of the reference path curve is evenly divided, so that the reference path points are evenly distributed on the reference path curve, the spacing between two adjacent reference path points in the longitudinal distance direction is equal, and the longitudinal distance coordinates of each reference path point form an arithmetic increase. Such a design, on the one hand, makes each segmentation position on the reference path curve the corresponding reference path point, and the reference path point can be obtained; on the other hand, the longitudinal distance of the reference path point can be quickly obtained by uniform division, reducing the amount of calculation. In other embodiments, the longitudinal distance of the reference path curve can be divided according to the preset division method to obtain the reference path point and its corresponding longitudinal distance coordinate.

[0065] Further, see Figure 4 , in the Frenet coordinate system, the reference path point p i The function expression of is formula (1):

[0066]

[0067] In formula (1), Indicates i reference waypoints; It represents the longitudinal distance coordinate of the reference path curve b along the tangent direction of the road reference curve in the Frenet coordinate system; s i Indicates the Frenet coordinate system i Reference waypoints p i Longitudinal distance coordinate along the tangent direction of the road reference curve; l It represents the lateral distance coordinate of the reference path curve b in the normal direction of the road reference curve in the Frenet coordinate system; l i Indicates the Frenet coordinate system i Reference waypoints p i The lateral distance coordinate in the normal direction of the road reference curve; Indicates the Frenet coordinate system i Reference waypoints p i The slope of the tangent line; That is d(d / ds) / ds , which means the Frenet coordinate system i Reference waypoints p i The derivative of the slope of the tangent line.

[0068] See also Figure 4 , according to the preset longitudinal distance difference Δ s, Uniformly divide the reference path curve to obtain several reference path points p i ,For example( s 0, l 0)、( s 2, l 2) ( s 4. l 4) and ( s 6. l 6). It can be understood that after uniform division, s 0 and s The distance between 2 is twice Δ s , s 2 and s The distance between 4 is twice Δ s, By analogy, the distance between every two adjacent reference path points along the tangent direction is equal, which is Δ s . Preset longitudinal distance difference Δ sThe larger it is, the fewer the number of reference path points will be, and vice versa.

[0069] It can be understood that based on the reference path curve b Figure 3 The Cartesian coordinate system and Figure 4 The mutual mapping relationship in the Frenet coordinate system, Figure 4 The 4 reference path points in p i ,exist Figure 3 The Cartesian coordinate system also has the corresponding reference path point position coordinates p i ( x i , y i ), x i Indicates the reference path point p i The horizontal distance coordinate along the x-axis in the Cartesian coordinate system; y i Indicates the reference path point p i The longitudinal distance coordinate along the y-axis in the Cartesian coordinate system. Figure 3 As shown, the coordinates of each reference path point include ( x 0, y 0)、( x 2, y 2) ( x 4. y 4) and ( x 6. y 6). That is, s 0, l 0) and ( x 0, y 0) corresponds to, ( s 2, l 2) and ( x 2, y 2) Corresponding, ( s 4. l 4) and ( x 4. y 4) Corresponding, ( s 6. l 6) and ( x 6. y 6) Correspondence.

[0070] Step S230, calculating and obtaining the lateral distance coordinates of the optimized path points according to the preset target path function and the longitudinal distance coordinates of the reference path points, wherein the preset target path function is constrained according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions.

[0071] In order to reduce the amount of calculation, in one embodiment, the number of optimized path points is the same as the number of reference path points. Further, in one embodiment, the longitudinal distance coordinate of each reference path point is the longitudinal distance coordinate of the corresponding optimized path point. The longitudinal distance coordinate of the optimized path point is s i Such a design can greatly reduce the computational load of the system and improve the efficiency of path planning.

[0072] Based on the longitudinal distance coordinates determined by the optimized path points, the lateral distance coordinates of the optimized path points are calculated according to the preset target path function and the longitudinal distance coordinates of the reference path points. It can be understood that by obtaining the lateral distance coordinates of the optimized path points and combining them with the longitudinal distance coordinates of the optimized path points, the clear position information of the optimized path points in the Frenet coordinate system is obtained. Such a design can facilitate the subsequent steps to plan the path curve according to the lateral distance coordinates and longitudinal distance coordinates of the optimized path points.

[0073] In order to optimize the cost of the preset target path function of the optimized path point, in one embodiment, the preset target path function is evaluated according to at least one of the following: the minimum cost of the distance between the planned path curve and the reference path curve, the cost of the planned path curve being close to the road reference curve, the second-order smoothness cost of the planned path curve, and the third-order smoothness cost of the planned path curve. Among them, the minimum cost of the distance between the planned path curve and the reference path curve reflects the degree of fit between the planned path curve and the reference path curve, that is, the degree of fit between each optimized path point and the corresponding reference path point. The smaller the distance between the planned path curve and the reference path curve, the closer the distance between each optimized path point and the corresponding reference path point. The longitudinal distance coordinates of each optimized path point and the corresponding reference path point are the same, so the smaller the distance between the planned path curve and the reference path curve, the smaller the difference between the longitudinal distance coordinates of each optimized path point and the corresponding reference path point. The cost of the planned path curve being close to the road reference curve reflects the degree of fit between the planned path curve and the road reference curve, each optimized path point and the corresponding reference path point. The smaller the cost of the planned path curve being close to the road reference curve, the closer the optimized path point is to the corresponding reference path point. The second-order smoothness cost of the planned path curve and the third-order smoothness cost of the planned path curve can both reflect the smoothness of the planned path curve. The optimized path point of the second-order smoothness cost of the planned path curve measures the smoothness of the optimized path point by the lateral distance coordinates of the three adjacent optimized path points, and the third-order smoothness cost of the planned path curve measures the smoothness of the optimized path point by the lateral distance coordinates of the four adjacent optimized path points. This embodiment performs cost evaluation on the preset target path function through smoothness costs of different orders, so that the cost evaluation is more accurate.

[0074] Furthermore, under the premise that the longitudinal distance coordinate of the optimized path point is determined, the preset target path function is constrained by the preset obstacle avoidance boundary condition and the preset curvature constraint condition, so as to constrain the lateral distance coordinate of the optimized path point, so as to obtain the optimized path point that makes the preset target path function reach the optimal value. In one embodiment, the preset curvature constraint condition is obtained according to the curvature function of the matching point of the road reference curve; wherein the matching point is obtained according to the corresponding mapping of the reference path point on the road reference curve. For ease of understanding, please refer to Figure 3 and Figure 4 ,based on Figure 4 The reference path curve in the Frenet coordinate system in i Reference waypoints p i, Can be Figure 3 The corresponding mapping is performed on the road reference curve to obtain the corresponding matching point p ri See also Figure 3 , matching point of road reference curve a p ri The function expression of is formula (2):

[0075]

[0076] In formula (2), ri Indicates the Cartesian coordinate system i Matching points; x ri Indicates the Cartesian coordinate system i The horizontal distance coordinates of the matching points along the x-axis; ri Indicates the Cartesian coordinate system i The longitudinal distance coordinates of the matching points along the y-axis; represents the direction angle of the reference path curve b in the Cartesian coordinate system; Indicates the Cartesian coordinate system i The direction angle of the matching points; Indicates i The curvature of the matching points; Indicates i Matching points p ri In one embodiment, the preset curvature constraint condition can be based on Obtain. It can be understood that the optimized path points are controlled by preset curvature constraints, so that the planned path curve generated according to the optimized path points can avoid the situation where the planned path curve obtained in the subsequent steps exceeds the limit of the car. In order to make the position of the optimized path points more reasonable and avoid the optimized path points exceeding the boundary of the road, in one embodiment, the obstacle avoidance boundary conditions include at least one of the following constraints: vehicle drivable boundary constraints, starting path point boundary constraints, ending path point boundary constraints and path point boundary constraints in the Frenet coordinate system. Among them, the vehicle drivable boundary constraints are used to limit the maximum and minimum values ​​of the lateral distance coordinates of the optimized path points, thereby limiting the driving boundaries of the car and ensuring that the planned path curve of the car will not deviate from the navigation path. Furthermore, by performing boundary constraints at the starting path point, the ending path point and each path point, each optimized path point from the starting path point to the ending path point is within the boundary constraint range. The following is a detailed description of the acquisition process of the lateral distance coordinates of the optimized path points.

[0077] In one embodiment, the specific process of calculating and obtaining the lateral distance coordinate of the optimized path point according to the preset target path function and the longitudinal distance coordinate is as follows:

[0078] Variables defining the lateral distance of the optimized path points ,in Represents the horizontal distance coordinate of the first optimized path point, Indicates the lateral distance coordinate of the second optimized path point; Indicates n The lateral distance coordinates of the optimized path points; n Indicates the number of optimized path points.

[0079] Define the preset target path function (3):

[0080]

[0081] In formula (3), Represents the minimum cost of the distance between the planned path curve and the reference path curve; The weight coefficient representing the minimum cost of the distance between the planned path curve and the reference path curve; Represents the cost of the planned path curve being close to the center line of the road, that is, the cost of the planned path curve being close to the reference curve of the road; The weight coefficient representing the cost of the planned path curve being close to the road reference curve; Represents the second-order smoothness cost of the planned path curve; The weight coefficient representing the second-order smoothness cost of the planned path curve; Represents the third-order smoothness cost of the planned path curve; The weight coefficient representing the third-order smoothness cost of the planned path curve; Representative i+ 1 lateral distance coordinate of a reference path point; Represents the distance between two adjacent optimized path points, that is, the distance between two adjacent path reference points.

[0082] Furthermore, the obstacle avoidance boundary conditions are set as shown in equations (4) to (13):

[0083] Among them, the vehicle drivable boundary constraint is set, that is, formula (4):

[0084]

[0085] In formula (4), Representative i+ The upper limit of the vehicle's drivability at 1 optimized path point, that is, the lateral distance coordinate of the upper limit of the vehicle's drivability, that is, the first i+ The minimum value of the horizontal distance coordinate of 1 optimized path point; Representative i+ 1 The lower limit of the vehicle's drivability at the optimized path point, that is, the lateral distance coordinate of the lower limit of the vehicle's drivability; that is, i+ The maximum value of the horizontal distance coordinate of an optimized path point.

[0086] Set the boundary constraints of the starting path point, that is, equations (5) to (7):

[0087]

[0088] Among them, in formula (5), Represents the lateral distance coordinate value of the starting point in the optimized path point in the Frenet coordinate system, that is, the lateral distance coordinate value of the first optimized path point.

[0089]

[0090] In formula (6), Represents the cost of the optimized path point at the starting point of the planned path curve being close to the road reference curve, that is, the cost of the first optimized path point being close to the road reference curve; Represents the first derivative of the starting point, that is, the first derivative of the optimized path curve at the starting point.

[0091]

[0092] In formula (7), Represents the second-order smoothness cost at the starting point of the planned path curve; Represents the second-order derivative of the starting point, that is, the second-order derivative of the optimized path curve at the starting point.

[0093] Set the boundary constraints of the end path point, that is, equations (8) to (10):

[0094]

[0095] In formula (8), Represents the lateral distance coordinate value of the end point in the optimized path point in the Frenet coordinate system, that is, n The horizontal distance coordinate value of the optimized path point.

[0096]

[0097] In formula (9), Represents the cost of the optimized path point at the end point of the planned path curve being close to the road reference curve, that is, n The cost of an optimized path point being close to the road reference curve; Represents the first derivative of the end point, that is, the first derivative of the optimized path curve at the end point.

[0098]

[0099] In formula (10), Represents the second-order smoothness cost at the end point of the planned path curve; Represents the second-order derivative of the end point, that is, the second-order derivative of the optimized path curve at the end point.

[0100] Set the path point boundary constraints, that is, equations (11) to (13):

[0101]

[0102] In formula (11), Represents the cost of the optimized path point of the planned path curve being close to the road reference curve, that is, i+ 1. The cost of optimizing the path point to be close to the road reference curve; Represents the minimum value of the first derivative of the lateral distance coordinate of the optimized path point; Represents the maximum value of the first derivative of the lateral distance coordinate of the optimized path point.

[0103]

[0104] In formula (12), Representative i+ 1. The second-order smoothness cost at the optimization path point; Represents the minimum value of the second derivative of the lateral distance coordinate of the optimized path point, where ddl Represents the second derivative of the lateral distance coordinate of the optimized path point; Represents the maximum value of the first derivative of the lateral distance coordinate of the optimized path point.

[0105]

[0106] In formula (13), Representative i+ 1. The third-order smoothness cost at an optimization path point; Represents the minimum value of the third derivative of the lateral distance coordinate of the optimized path point, where ddl represents the third derivative of the lateral distance coordinate of the optimized path point; Represents the maximum value of the third derivative of the lateral distance coordinate of the optimized path point.

[0107] Set the preset curvature constraint conditions, that is, equations (14) to (15):

[0108]

[0109] In formula (14), Representative i+ 1 optimizes the curvature at the path points, Representative i+ Curvature at 1 matching point.

[0110]

[0111] In formula (15), Represents the lower boundary of the curvature that the vehicle can execute, that is, represents the minimum value of the curvature that the vehicle can execute; Represents the upper limit of the curvature that the vehicle can execute, that is, represents the maximum value of the curvature that the vehicle can execute.

[0112] Therefore, through the preset target path function, obstacle avoidance boundary conditions and preset curvature constraint conditions of equations (3) to (15), the optimized path points can be constrained to prevent the optimized path points from exceeding the boundaries. At the same time, the curvature of the planned path curve obtained by the optimized path points can be indirectly controlled by constraining the lateral distance of the optimized path points.

[0113] Step S240, obtaining a planned path curve according to the lateral distance coordinates of the optimized path points and the longitudinal distance coordinates of the reference path points.

[0114] Since several reference path points are obtained by uniformly dividing the reference path curve, and the longitudinal distance coordinates of the reference path points are arithmetically increasing, when the longitudinal distance coordinate of one of the reference path points is determined, the longitudinal distance coordinates of other reference path points can be obtained, and the optimized path point corresponds to the reference path point, that is, the longitudinal distance coordinate of the optimized path point is equal to the longitudinal distance coordinate of the reference path point. Therefore, by obtaining the lateral distance coordinate of the optimized path point, combined with the longitudinal distance coordinate that is the same as the corresponding reference path point, the position information corresponding to a single optimized path point in the Frenet coordinate system, that is, the coordinate value, can be obtained. By passing through each optimized path point, a planned path curve that avoids exceeding the limit of the car and is reasonable and drivable can be obtained. In other embodiments, the planned path curve may also pass through some of the optimized path points, or may be close to the optimized path points without passing through the optimized path points.

[0115] See also Figure 5 According to the above steps, the planned path curve c of the present application is obtained. The planned path curve c is generated between the road reference curve a and the reference path curve b. The curvature of the planned path curve c is smaller than that of the reference path curve b, and is more suitable for car driving. Figure 6 , the change curve t_theta of the direction angle of the unoptimized path curve, i.e. the reference path curve, has a larger change range than the change curve theta of the direction angle of the planned path curve c of the present application, indicating that the path planning method of the present application can effectively constrain the change of the direction angle of the planned path curve c, and the change of the direction angle also reflects the change of the curvature, thus indicating that the path planning method of the present application can effectively constrain the curvature of the planned path curve c. Figure 7 The curvature change curve t_kappa of the unoptimized path curve, i.e., the reference path curve, has a larger range of variation than the curvature change curve kappa of the planned path curve c of the present application, wherein the curvature of the planned path curve c is greater than or equal to 0.2, and the curvature of the unoptimized path curve is less than 0.2, which more intuitively illustrates the constraint of the present application on the curvature of the planned path curve c.

[0116] In summary, the path planning method of the present application establishes a Frenet coordinate system based on the center line of the road in the Cartesian coordinate system, that is, the road reference curve. According to the relative position of the reference path curve in the Cartesian coordinate system and the road reference curve, a corresponding reference path curve is set in the Frenet coordinate system, and the longitudinal distance of the reference path curve is evenly cut, so as to fix the difference between the longitudinal distance coordinates of the two adjacent reference path points on the reference path curve, and then fix the difference between the longitudinal distance coordinates of the two adjacent optimized path points on the planning path curve, and then the longitudinal distance coordinates of each optimized path point can be determined, which reduces the amount of calculation. The position information of the optimized path point can be determined by calculating and obtaining the lateral distance coordinates of the optimized path point. The lateral distance coordinates of the optimized path point are calculated by the preset target path function and the longitudinal distance coordinates. The preset target path function can constrain the lateral distance coordinates of the optimized path point through obstacle avoidance boundary conditions including the starting path point boundary constraint, the ending path point boundary constraint, the path point boundary constraint and the vehicle drivable boundary constraint, so as to avoid the optimized path point from exceeding the boundary. The preset target path function can also constrain the curvature of the optimized path points through preset curvature constraints to prevent the curvature of the optimized path points from exceeding the upper and lower boundaries of the curvature that the vehicle can execute, making the planned path curve obtained based on the optimized path points more reasonable and feasible.

[0117] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides a path planning device, a car, a non-temporary machine-readable storage medium and corresponding embodiments.

[0118] See also Figure 3 The present application provides a path planning device 30 according to an embodiment, which includes:

[0119] The reference curve acquisition module 310 is used to acquire a reference path curve.

[0120] The reference path point acquisition module 320 is used to acquire reference path points of the reference path curve.

[0121] The optimized path point acquisition module 330 is used to obtain the position information of the optimized path point according to the position information of the reference path point and the preset target path function; wherein the preset target path function is constrained according to the preset obstacle avoidance boundary condition and the preset curvature constraint condition.

[0122] The planned path acquisition module 340 is used to obtain the planned path curve according to the position information of the optimized path points.

[0123] In order to obtain the reference path curve, the reference curve acquisition module 310 uses the center line of the road as the road reference curve in the Cartesian coordinate system; and obtains the reference path curve in the Frenet coordinate system according to the road reference curve. The reference path point acquisition module 320 evenly divides the longitudinal distance of the reference path curve in the Frenet coordinate system, collects the reference path points of the reference path curve, and obtains the longitudinal distance coordinates corresponding to the reference path points. The optimized path point acquisition module 330 uses the longitudinal distance coordinates corresponding to the reference path points as the longitudinal distance coordinates of the corresponding optimized path points, and calculates the lateral distance coordinates of the optimized path points according to the preset target path function and the longitudinal distance coordinates of the reference path points, thereby acquiring the optimized path points. The preset target function can constrain the position of the optimized path points according to the preset obstacle avoidance boundary conditions and the preset curvature constraint conditions to avoid the optimized path points from exceeding the boundaries, and can also control the curvature of the curve formed by the optimized path points to avoid the curvature of the curve obtained according to the optimized path points from exceeding the limit of the car. The planned path acquisition module 340 obtains the planned path curve according to the lateral distance coordinates of the optimized path points and the longitudinal distance coordinates of the reference path points.

[0124] In summary, the path planning device 30 of the present application can obtain a reference path curve according to the reference curve acquisition module 310, the reference path point acquisition module 320 can collect reference path points of the reference path curve and obtain the position information of the reference path points, the optimization path point acquisition module 330 can obtain the position information of the optimization path points according to the position information of the reference path points and the preset target path function, and the planning path acquisition module 340 can obtain the planning path curve according to the position information of the optimization path points. Such a design can obtain the position information of the optimization path points by constraining the preset curvature constraint conditions through the preset target path function, and then obtain the planning path curve, so as to avoid the curvature of the planning path curve exceeding the vehicle limit and provide a reasonable path that the vehicle can drive.

[0125] The present application also provides a car, the car of the present application includes a path planning device of any of the above embodiments. The car travels according to the planned path curve generated by the path planning device, thereby ensuring that the car travels on a safe and executable driving path constrained by curvature.

[0126] The present application also provides an electronic device, the electronic device comprising a memory and a processor. The memory stores executable code, and when the executable code is executed by the processor, the processor executes the path planning method described in any of the above embodiments.

[0127] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0128] The memory may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage systems. Among them, ROM can store static data or instructions required by the processor or other modules of the computer. The permanent storage system may be a readable and writable storage system. The permanent storage system may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage system uses a large-capacity storage system (such as a magnetic or optical disk, flash memory) as a permanent storage system. In other embodiments, the permanent storage system may be a removable storage device (such as a floppy disk, an optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0129] The memory stores executable codes, and when the executable codes are processed by the processor, the processor can execute part or all of the methods described above.

[0130] The non-temporary machine-readable storage medium of the present application stores executable code thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the above method.

[0131] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0132] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0133] Alternatively, the present application can also be implemented as a non-temporary machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or electronic device, server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0134] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A path planning method, Features: Obtain a reference path curve; Collecting reference path points of the reference path curve; including: uniformly dividing the longitudinal distance along the reference path curve in the Frenet coordinate system to obtain the longitudinal distance coordinates of the reference path points; According to the position information of the reference path point and the preset target path function, the position information of the optimized path point is obtained; which includes: according to the preset target path function and the longitudinal distance coordinate of the reference path point, the lateral distance coordinate of the optimized path point is calculated; according to the lateral distance coordinate and the longitudinal distance coordinate of the reference path point, the position information of the optimized path point is obtained; wherein, the preset target path function is constrained according to the preset obstacle avoidance boundary condition and the preset curvature constraint condition on the premise that the longitudinal distance coordinate of the optimized path point is determined, so as to constrain the lateral distance coordinate of the optimized path point; the obstacle avoidance boundary condition includes at least one of the following constraints: the vehicle drivable boundary constraint in the Frenet coordinate system, the starting path point boundary constraint, the ending path point boundary constraint and the path point boundary constraint; A planned path curve is obtained according to the position information of the optimized path points.

2. The path planning method according to claim 1, It is characterized in that The obtaining of the reference path curve comprises: In the Cartesian coordinate system, the center line of the road is taken as the road reference curve; The reference path curve is obtained in a Frenet coordinate system according to the road reference curve.

3. The path planning method according to claim 1, It is characterized in that The preset target path function is evaluated according to at least one of the following: The minimum cost of the distance between the planned path curve and the reference path curve, the cost of the planned path curve being close to the road reference curve, the second-order smoothness cost of the planned path curve, and the third-order smoothness cost of the planned path curve.

4. The path planning method according to claim 2, It is characterized in that The preset curvature constraint condition is obtained according to the curvature function of the matching point of the road reference curve; wherein the matching point is obtained according to the corresponding mapping of the reference path point on the road reference curve.

5. A path planning device, It is characterized in that include: A reference curve acquisition module, used to acquire a reference path curve; A reference path point acquisition module, used for acquiring reference path points of the reference path curve; The method includes: uniformly dividing the longitudinal distance along the reference path curve in the Frenet coordinate system to obtain the longitudinal distance coordinates of the reference path points; The optimized path point acquisition module is used to obtain the position information of the optimized path point according to the position information of the reference path point and the preset target path function; including: calculating and obtaining the lateral distance coordinate of the optimized path point according to the preset target path function and the longitudinal distance coordinate of the reference path point; obtaining the position information of the optimized path point according to the lateral distance coordinate and the longitudinal distance coordinate of the reference path point; wherein, the preset target path function is constrained according to the preset obstacle avoidance boundary condition and the preset curvature constraint condition on the premise that the longitudinal distance coordinate of the optimized path point is determined, so as to constrain the lateral distance coordinate of the optimized path point; the obstacle avoidance boundary condition includes at least one of the following constraints: vehicle drivable boundary constraint, starting path point boundary constraint, ending path point boundary constraint and path point boundary constraint in the Frenet coordinate system; The planned path acquisition module is used to obtain the planned path curve according to the position information of the optimized path points.

6. A car, It is characterized in that Includes the path planning device as described in claim 5.

7. A non-transitory machine-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to execute the method according to any one of claims 1 to 4.