Driving method and device, electronic equipment and computer readable storage medium

Abstract

The application discloses a driving method and device, electronic equipment and computer readable storage medium, and belongs to the technical field of automatic driving. The method comprises the following steps: obtaining a first path based on the driving data of a vehicle in a Frenet coordinate system; converting the first path in the Frenet coordinate system into a second path in a Cartesian coordinate system; optimizing the second path according to an optimization strategy based on the Cartesian coordinate system to obtain a target path; and controlling the movement of the vehicle based on the target path. The application combines the advantages of the Frenet coordinate system, i.e. the complexity of path planning can be simplified, and the advantages of the Cartesian coordinate system, i.e. the kinematic constraints and curvature constraints of the vehicle can be considered, to obtain a more accurate target path and reduce the dependence on the smoothness of a reference line.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and more particularly, to a driving method and device, an electronic device, and a computer readable storage medium. BACKGROUND

[0002] The Frenet coordinate system is a dynamic coordinate system, and the Frenet coordinate system represents the position of a vehicle on a road through a longitudinal distance and a transverse distance. Although path planning in the Frenet coordinate system can simplify the motion planning problem, local planning in the Frenet coordinate system may not meet the actual kinematic requirements of the vehicle, causing the vehicle to be unable to travel along the planned path and easily causing traffic accidents. SUMMARY

[0003] The present application provides a driving method and device, an electronic device, and a computer readable storage medium to improve the above-mentioned defects.

[0004] In a first aspect, an embodiment of the present application provides a driving method, which comprises: obtaining a first path in a Frenet coordinate system based on acquired driving data of a vehicle; converting the first path in the Frenet coordinate system into a second path in a Cartesian coordinate system; optimizing the second path according to an optimization strategy based on the Cartesian coordinate system to obtain a target path; and controlling the motion of the vehicle based on the target path.

[0005] Optionally, for a possible implementation, the optimization strategy includes a target function and a constraint condition, and the optimization of the second path according to the optimization strategy based on the Cartesian coordinate system to obtain the target path includes: determining a plurality of trajectory points on the second path as reference path points; inputting the position information of the reference path points into the target function for solving to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies the constraint condition; and determining the target path according to the position information of a plurality of predicted path points.

[0006] Optionally, for a possible implementation, the target function includes an adjacent frame control parameter, and the adjacent frame control parameter represents the difference between first data and second data, wherein the first data represents the third derivative of the current frame predicted path point with respect to the arc length, the second data represents the third derivative of the last frame predicted path point with respect to the arc length, and the arc length is a smooth curve between two adjacent path points.

[0007] Optionally, for a possible implementation, the target function is: wherein, and state variables representing a target function, control variables representing the target function, wherein x k , y k represent the horizontal and vertical coordinates of a predicted path point, respectively, represent the first, second and third derivatives of the horizontal coordinate of the predicted path point with respect to the arc length, respectively, represents the third derivative of the horizontal coordinate of the previous predicted path point, represent the first, second and third derivatives of the vertical coordinate of the predicted path point with respect to the arc length, respectively, represents the third derivative of the vertical coordinate of the previous predicted path point; wherein, represent the horizontal and vertical coordinates of a reference path point, respectively, represents the first derivative of the horizontal coordinate of the reference path point with respect to the arc length, represents the first derivative of the vertical coordinate of the reference path point with respect to the arc length, κ k represents the curvature of the predicted path point, represents the curvature of the reference path point; wherein ω1, ω2, ω3, ω4, ω5, ω6 are preset weights, N is the number of predicted path points, and k is a serial number.

[0008] Optionally, for a possible implementation, the constraint condition comprises a curvature constraint, and before the position information of the reference path point is input into the target function for solving to obtain the predicted path point corresponding to each reference path point, the method further comprises: obtaining a preliminary curvature constraint based on the reference path point, and wherein, based on and determining the preliminary curvature constraint as: wherein κ max is a maximum curvature limit; the preliminary curvature constraint is solved by a primal-dual interior point method to obtain a determined λ k1 and μ k1 ; and the curvature constraint is determined according to λ k1 and μ k1 :

[0009] Optionally, for a possible implementation, the constraint condition comprises a curvature variation constraint, and before the position information of the reference path point is input into the target function for solving to obtain the predicted path point corresponding to each reference path point, the method further comprises: obtaining a preliminary curvature variation constraint based on the reference path point, and wherein, based on and The preliminary curvature change constraint is determined as follows: wherein Δκ max is the maximum curvature change limit; the preliminary curvature change constraint is solved by the original dual interior point method to obtain the determined λ k1 , μ k1 , λ k1+1 and μ k1+1 ; the curvature change constraint is determined according to λ k1 , μ k1 , λ k1+1 and μ k1+1 .

[0010] Optionally, for a possible implementation, the constraint condition includes an obstacle constraint, and before the position information of the reference path point is input into the objective function for solving to obtain the predicted path point corresponding to each reference path point, the method further includes: obtaining a plurality of first trajectory points based on a global path, the global path being a planned path of the vehicle from a starting point to an ending point; determining boundary points of a normal direction of the first trajectory point based on position information of an obtained obstacle, the normal direction being perpendicular to a reference line on which the first trajectory point is located; connecting same-side boundary points of adjacent two first trajectory points and connecting left and right boundary points of the first trajectory point to obtain a plurality of convex polygons; combining the plurality of convex polygons into a preset position region; and establishing the obstacle constraint based on the preset position region.

[0011] Optionally, for a possible implementation, the second path is optimized according to the optimization strategy based on the Cartesian coordinate system to obtain a target path, including:

[0012] determining whether the second path satisfies a preset condition; and if the second path does not satisfy the preset condition, optimizing the second path according to the optimization strategy based on the Cartesian coordinate system to obtain a target path.

[0013] In a second aspect, the embodiments of the present application further provide a driving device, including: a conversion unit, configured to obtain a first path based on obtained driving data of a vehicle in a Frenet coordinate system; a construction unit, configured to convert the obtained first path in the Frenet coordinate system into a second path in a Cartesian coordinate system; a determination unit, configured to optimize the second path according to an optimization strategy based on the Cartesian coordinate system to obtain a target path; and a control unit, configured to control movement of the vehicle based on the target path.

[0014] In a third aspect, the embodiments of the present application further provide an electronic device, comprising: one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the memory, the one or more application programs are configured to be executed by the one or more processors, and the one or more application programs are configured to execute the method described above.

[0015] In a fourth aspect, the embodiments of the present application further provide a computer readable medium, wherein the readable storage medium stores a program code executable by a processor, and the program code, when executed by the processor, causes the processor to execute the method described above.

[0016] The driving method and device, the electronic device and the computer readable storage medium provided by the present application first obtain a first path based on the driving data of the vehicle in the Frenet coordinate system; convert the first path in the Frenet coordinate system into a second path in the Cartesian coordinate system; then optimize the second path according to an optimization strategy based on the Cartesian coordinate system to obtain a target path; and finally control the movement of the vehicle based on the target path.

[0017] In the Frenet coordinate system, the position and dynamics of the vehicle can be described as longitudinal offset and lateral offset along the reference line, which can simplify the complexity of path planning. However, the effectiveness and accuracy of this representation method largely depend on the smoothness of the reference line. When the reference line is relatively smooth, the accuracy of the first path obtained is relatively high. However, when the reference line is not smooth, the accuracy of the first path obtained is relatively low. The reference line is provided by a global path, which is a pre-acquired planning path of the vehicle from the starting point to the end point.

[0018] Therefore, the present application converts the first path in the Frenet coordinate system into a second path in the Cartesian coordinate system, and then optimizes the second path based on the optimization strategy of the Cartesian coordinate system to obtain a target path. Compared with the first path in the Frenet coordinate system, the target path obtained by the present application has higher accuracy and is more consistent with the kinematic model of the vehicle.

[0019] Other features and advantages of the embodiments of the present application will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the embodiments of the present application. The objects and other advantages of the embodiments of the present application will be realized and achieved by means of the structures particularly pointed out in the written description and claims, and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0021] Figure 1 The method flow chart of driving is shown, which is provided by the embodiments of the present application;

[0022] Figure 2 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0023] Figure 3 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0024] Figure 4 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0025] Figure 5 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0026] Figure 6 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0027] Figure 7 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0028] Figure 8 The method flow chart of driving is shown, which is provided by another embodiment of the present application;

[0029] Figure 9 The schematic diagram of convex polygon is shown, which is provided by the embodiments of the present application;

[0030] Figure 10 The schematic diagram of preset position area is shown, which is provided by the embodiments of the present application;

[0031] Figure 11 The schematic diagram of target path is shown, which is provided by the embodiments of the present application;

[0032] Figure 12 The structural block diagram of driving device is shown, which is provided by the embodiments of the present application;

[0033] Figure 13 The structural block diagram of electronic device is shown, which is provided by the embodiments of the present application;

[0034] Figure 14 The structural block diagram of computer readable storage medium is shown, which is provided by the embodiments of the present application. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. The components of the embodiments of the present application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are within the scope of protection of the present application.

[0036] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0037] Although path planning in the Frenet coordinate system can simplify motion planning problems, local planning in the Frenet coordinate system may not meet the actual kinematic requirements of the vehicle, causing the vehicle to fail to travel along the planned path, which can easily lead to traffic accidents.

[0038] Therefore, this application provides a driving method, apparatus, electronic device, and computer-readable storage medium to solve or partially solve the above-mentioned problems.

[0039] Please see Figure 1 , Figure 1 A flowchart of a driving method provided in an embodiment of this application is shown, which specifically includes steps S110 to S140.

[0040] Step S110: In the Frenet coordinate system, based on the acquired vehicle driving data, obtain the first path.

[0041] The Frenet coordinate system is a dynamic coordinate system commonly used for vehicle motion planning on roads. The Frenet coordinate system uses two parameters, longitudinal distance and lateral distance, to represent the position of a vehicle on the road.

[0042] The longitudinal distance refers to a distance along a road reference line, and the longitudinal distance is measured from a fixed starting point and increases as the vehicle moves along the road. The lateral distance refers to a distance perpendicular to the road reference line, and can represent an offset of the vehicle relative to the reference line. The road reference line can be a road center line or a lane line.

[0043] Further, the driving data of the vehicle can include data collected by the vehicle during driving.

[0044] For example, the driving data of the vehicle can include road information, vehicle state information, surrounding environment information, traffic information, and weather information, etc. The road information can include road geometry, road reference line, and road markings and traffic signs. The vehicle information can include vehicle position information, speed information, acceleration information, and steering angle information, etc. The surrounding environment information can include obstacle information and positions and speeds of other vehicles. The traffic information can include real-time traffic information and traffic signal rules.

[0045] Since path planning in the Frenet coordinate system can simplify the motion planning problem, a first path can be obtained in the Frenet coordinate system based on the obtained driving data of the vehicle.

[0046] The first path is a local planning path obtained by path planning in the Frenet coordinate system based on the driving data. In the Frenet coordinate system, the first path can be a smooth curve composed of a plurality of trajectory points connected in sequence.

[0047] In the embodiments of the present application, the method is applied to an automatic driving scenario, the driving data is data collected by the vehicle in the automatic driving scenario, and the first path is a driving route planned for the vehicle in the Frenet coordinate system based on the existing driving data of the vehicle.

[0048] Step S120: converting the obtained first path in the Frenet coordinate system into a second path in the Cartesian coordinate system.

[0049] Since it is difficult to consider the vehicle kinematic model, the trajectory point curvature and the curvature change rate constraint in the Frenet coordinate system, especially in a large-curvature curve, the accuracy of the local planning in the Frenet coordinate system is low. Moreover, path planning in the Cartesian coordinate system can realize the trajectory point curvature constraint, the curvature change constraint, and the vehicle kinematic constraint.

[0050] Therefore, the present application converts the obtained first path in the Frenet coordinate system into a second path in the Cartesian coordinate system, so as to optimize the second path in the Cartesian coordinate system and improve the accuracy of predicting the second path.

[0051] The first path in the Frenet coordinate system can be converted into a second path in the Cartesian coordinate system through a mapping relationship between the Frenet coordinate system and the Cartesian coordinate system.

[0052] The optimization strategy can include a target function and a constraint condition, and details are described in subsequent embodiments.

[0053] Step S130: optimizing the second path according to an optimization strategy based on the Cartesian coordinate system to obtain a target path.

[0054] It should be noted that the curvature and curvature change parameters obtained in the Frenet coordinate system are not accurate, so it is difficult to realize curvature optimization and curvature change optimization in the Frenet coordinate system.

[0055] The Cartesian coordinate system usually corresponds to an optimization strategy, which includes curvature optimization, curvature change optimization, and obstacle optimization of a path input into the Cartesian coordinate system, etc. In the embodiments of the present application, at least part of the optimization means supported by the Cartesian coordinate system can be determined as the optimization strategy of the present application, or the optimization means supported by the Cartesian coordinate system can be improved to obtain the optimization strategy of the present application. Details are described in subsequent embodiments.

[0056] The related parameters of the second path are input into the optimization strategy, and the target path is obtained by optimizing the optimization strategy. The target path is a planning path obtained by optimizing the second path.

[0057] In an optional embodiment, the optimization strategy is solved, the optimized strategy is converted into an incremental form, a quadratic programming problem is constructed, and the solution is solved by an operator splitting quadratic program (OSQP).

[0058] Specifically, a plurality of reference path points can be obtained according to the second path, the related parameters of the reference path points are input into the optimization strategy, the reference path points are optimized through the optimization strategy, and a predicted path point corresponding to each reference path point is obtained. The predicted path point satisfies the constraint condition, and then the target path is obtained based on the predicted path point. Details are described in subsequent embodiments.

[0059] Step S140: controlling vehicle motion based on the target path.

[0060] When the target path is obtained, the vehicle motion can be controlled based on the target path, so that the vehicle can more smoothly and accurately reach the destination.

[0061] In the embodiment of the present application, a driving method is provided. In the automatic driving scenario of a vehicle, first, in the Frenet coordinate system, a first path is obtained based on the acquired driving data of the vehicle; the acquired first path in the Frenet coordinate system is converted into a second path in the Cartesian coordinate system; then, an optimization strategy is determined based on the Cartesian coordinate system; second, the second path is optimized according to the optimization strategy to obtain a target path; finally, the vehicle motion is controlled based on the target path.

[0062] The present application converts the first path in the Frenet coordinate system into the second path in the Cartesian coordinate system, and then optimizes the second path based on the optimization strategy of the Cartesian coordinate system to obtain the target path. Compared with the first path in the Frenet coordinate system, the target path obtained by the present application has higher accuracy and is more consistent with the kinematic model of the vehicle.

[0063] Although the position and dynamics of the vehicle can be described as longitudinal offset and lateral offset along the reference line in the Frenet coordinate system, which can simplify the complexity of path planning, the effectiveness and accuracy of this representation method depend largely on the smoothness of the reference line. When the reference line is relatively smooth, the accuracy of the obtained first path is relatively high, but when the reference line is not smooth, the accuracy of the obtained first path is relatively low.

[0064] The reference line is a planned route provided by a global path, and the global path is a pre-acquired planned path of the vehicle from a starting point to an ending point.

[0065] Since the Cartesian coordinate system usually corresponds to an optimization strategy, the optimization strategy includes curvature optimization, curvature change optimization, obstacle optimization and the like for the path input into the Cartesian coordinate system. The optimization strategy of the present application is at least part of the optimization means supported by the Cartesian coordinate system. When the reference line is not smooth, the accuracy of the obtained second path is relatively low. The present application can optimize the second path through the optimization strategy to obtain an optimized target path. The present application combines the advantage of simplifying the complexity of path planning in the Frenet coordinate system and the advantage of the optimization strategy corresponding to the Cartesian coordinate system to obtain a more accurate target path, thereby reducing the dependence on the smoothness of the reference line.

[0066] Please refer to Figure 2 , Figure 2 A method flowchart of a driving method provided by an embodiment of the present application is shown, which specifically includes steps S210 to S260.

[0067] Step S210: In the Frenet coordinate system, a first path is obtained based on the acquired driving data of the vehicle.

[0068] Step S220: converting the first path in the Frenet coordinate system into a second path in the Cartesian coordinate system.

[0069] It should be noted that the optimization strategy includes a target function and a constraint condition.

[0070] It should be noted that the optimization strategy includes a target function and a constraint condition.

[0071] Steps S210 and S220 have been described in detail in the foregoing embodiments, and will not be described here.

[0072] The second path is a continuous curve in the Cartesian coordinate system, and the trajectory points are discrete points on the second path. The second path can be used to obtain a plurality of trajectory points as reference path points.

[0073] It should be noted that the way of obtaining a plurality of trajectory points from the second path can include a plurality of ways.

[0074] In an optional embodiment, a plurality of trajectory points on the second path are obtained as reference path points by curve length, and the curve length between adjacent two reference points is the same.

[0075] For example, the total length of the second path is 10 meters, 500 reference path points are obtained from the curve where the second path is located, and the length of the curve between adjacent two reference path points is 0.02 meters. The reference path points are obtained by obtaining the coordinates of the reference path points in the Cartesian coordinate system, wherein the coordinates of the reference path points include horizontal coordinates and vertical coordinates.

[0076] In an optional embodiment, a plurality of trajectory points on the second path are obtained as reference path points by curvature. On a curve with small curvature, the curve length between adjacent two reference path points is longer, and on a curve with large curvature, the curve length between adjacent two reference path points is shorter.

[0077] For example, the total length of the second path is 10 meters, 500 reference path points are obtained from the curve where the second path is located, and the length of the curve between adjacent two reference path points is 0.02 meters. The reference path points are obtained by obtaining the coordinates of the reference path points in the Cartesian coordinate system, wherein the coordinates of the reference path points include horizontal coordinates and vertical coordinates.

[0078] Since fewer reference path points can accurately represent the path trajectory of the curve with small curvature, and more reference path points are needed to accurately represent the path trajectory of the curve with large curvature, the application obtains a plurality of trajectory points on the second path as reference path points by curvature. The obtained reference path points can more accurately represent the second path, and the second path can be accurately optimized.

[0079] Step S240: inputting the position information of the reference path point into the objective function to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies the constraint condition.

[0080] The position information of the reference path point is inputted into the objective function to obtain a predicted path point corresponding to each reference path point, and the predicted path point satisfies the constraint condition, that is, through solving the objective function, the coordinates of the predicted path point satisfying all the constraint conditions are obtained, and the predicted path point is more consistent with the vehicle kinematics model than the reference path point.

[0081] Step S250: determining a target path according to the position information of the plurality of predicted path points.

[0082] Specifically, the plurality of predicted path points are connected by a smooth curve in sequence, and then the connected curve is taken as the target path, and the target path is a local planning path of the vehicle.

[0083] Step S260: controlling the vehicle motion based on the target path.

[0084] The step S260 has been described in detail in the foregoing embodiments, and will not be described here.

[0085] The embodiments of the present application determine the reference path point based on the second path, input the reference path point into the objective function to obtain a predicted path point corresponding to each reference path point, and then obtain the target path according to the predicted path point, thereby realizing the optimization of the second path through the objective function and the constraint condition, and obtaining the target path with higher accuracy and consistent with the kinematics model.

[0086] Please refer to Figure 3 , Figure 3 A method flowchart of a driving method provided by the embodiments of the present application is shown, and the method specifically includes steps S310 to S360.

[0087] Step S310: obtaining a first path based on the acquired driving data of the vehicle in a Frenet coordinate system.

[0088] Step S320: converting the acquired first path in the Frenet coordinate system into a second path in a Cartesian coordinate system.

[0089] Step S330: determining a plurality of trajectory points on the second path as reference path points.

[0090] The steps S310 and S330 have been described in detail in the foregoing embodiments, and will not be described here.

[0091] Step S340: input the position information of the reference path point into the objective function for solving to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies the constraint condition. The optimization strategy includes an objective function and a constraint condition, the objective function includes adjacent frame control parameters, the adjacent frame control parameters represent the difference between the first data and the second data, wherein the first data represents the third derivative of the current frame predicted path point with respect to the arc length, the second data represents the third derivative of the last frame predicted path point with respect to the arc length, and the arc length is a smooth curve between two adjacent path points.

[0092] Specifically, in the process of vehicle driving, the collected driving data is updated in real time, the driving data collected at the last time is optimized at the last time to obtain the predicted path point of the last frame, and the driving data collected at the current time is optimized at the current time to obtain the predicted path point of the current frame.

[0093] The third derivative of the current frame predicted path point with respect to the arc length represents the acceleration change rate of the current frame predicted path point, which can also be understood as the control amount of the current frame predicted path point, and the third derivative of the last frame predicted path point with respect to the arc length represents the acceleration change rate of the last frame predicted path point, which can also be understood as the control amount of the last frame predicted path point. The adjacent frame control parameters represent the difference between the control amount of the current frame predicted path point and the control amount of the last frame predicted path point, and the smaller the adjacent frame control parameters, the smaller the change of the control amount of the adjacent two frames of path optimization, and the more similar the paths between the two frames. The difference between the adjacent frame control parameters can be constrained by the objective function to reduce the path jump between multiple frames and improve the stability of vehicle driving. Further, the objective function is: wherein, and represent the state quantity of the objective function, represent the control quantity of the objective function,

[0094] wherein, x k , y k respectively represent the horizontal coordinate and the vertical coordinate of the predicted path point, respectively represent the first derivative, the second derivative and the third derivative of the horizontal coordinate of the predicted path point with respect to the arc length, represent the third derivative of the horizontal coordinate of the last predicted path point, respectively represent the first derivative, the second derivative and the third derivative of the vertical coordinate of the predicted path point with respect to the arc length, represent the third derivative of the vertical coordinate of the last predicted path point.

[0095] wherein, respectively represent the horizontal coordinate and the vertical coordinate of the reference path point, represents the first derivative of the horizontal coordinate of the reference path point with respect to the arc length, represents the first derivative of the vertical coordinate of the reference path point with respect to the arc length, κ k represents the curvature of the predicted path point, represents the curvature of the reference path point.

[0096] wherein, ω1, ω2, ω3, ω4, ω5, ω6 are preset weights, N is the number of predicted path points, and k is the serial number.

[0097] It should be noted that, can be understood as the orientation of the kth predicted path point in the horizontal coordinate direction, can be understood as the acceleration of the kth predicted path point in the horizontal coordinate direction, can be understood as the acceleration change rate of the kth predicted path point in the horizontal coordinate direction. can be understood as the acceleration change rate of the kth predicted path point in the horizontal coordinate direction of the previous frame.

[0098] Similarly, can be understood as the orientation of the kth predicted path point in the vertical coordinate direction, can be understood as the acceleration of the kth predicted path point in the vertical coordinate direction, can be understood as the acceleration change rate of the kth predicted path point in the vertical coordinate direction. can be understood as the acceleration change rate of the kth predicted path point in the vertical coordinate direction of the previous frame.

[0099] Therefore, represents the square of the distance between the reference path point and the predicted path point, represents the square of the difference between the orientation of the reference path point and the orientation of the predicted path point, represents the square of the difference between the curvature of the reference path point and the curvature of the predicted path point, in the process of solving the objective function, it is expected that the predicted path point is as close as possible to the reference path point, the orientation at the predicted path point is as similar as possible to the orientation at the reference path point, and the curvature of the predicted path point is as similar as possible to the curvature of the reference path point.

[0100] wherein, represents the square sum of the horizontal acceleration and the vertical acceleration of the predicted path point, The square sum of the acceleration rate of the predicted path point in the horizontal coordinate direction and the acceleration rate in the vertical coordinate direction is represented. In the process of solving the objective function, the acceleration of the predicted path point is expected to be as small as possible, and the acceleration rate of the predicted path point is expected to be as small as possible, thereby improving the stability of vehicle driving and improving the comfort of passengers.

[0101] wherein, The square of the difference between the acceleration rate of the predicted path point in the current frame and the acceleration rate of the predicted path point in the previous frame is represented. In the process of solving the objective function, the acceleration rate of the predicted path point in the current frame is expected to be as similar as possible to the acceleration rate of the predicted path point in the previous frame, thereby ensuring the consistency of the control amount in the current frame and the control amount in the previous frame, and improving the stability of vehicle driving.

[0102] Step S350: determining a target path according to position information of a plurality of predicted path points.

[0103] Step S360: controlling vehicle movement based on the target path.

[0104] Wherein, step S330 and step S360 have been described in detail in the foregoing embodiments, and will not be repeated here.

[0105] Please refer to Figure 4 , Figure 4 A method flowchart of a driving method provided by an embodiment of the application is shown, and the method specifically includes steps S410 to S4100.

[0106] Step S410: obtaining a first path based on the driving data of the vehicle in the Frenet coordinate system.

[0107] Step S420: converting the first path in the Frenet coordinate system into a second path in the Cartesian coordinate system.

[0108] Wherein, step S410 and step S420 have been described in detail in the foregoing embodiments, and will not be repeated here.

[0109] It should be noted that the optimization strategy includes an objective function and a constraint condition.

[0110] In the actual driving process of the vehicle, the curvature of the curve path actually walked by the vehicle in turning is limited. Therefore, the curvature constraint condition needs to be considered when planning the path.

[0111] The constraint condition of the application includes a curvature constraint, which is used to constrain the predicted path point obtained by the objective function to satisfy the curvature constraint range, thereby making the accuracy of the target path obtained higher. The detailed content of the objective function is described in the foregoing embodiments, and will not be repeated here.

[0112] Step S430: obtaining the preliminary curvature constraint based on the reference path point and wherein,

[0113] Step S440: determining the preliminary curvature constraint based on and The preliminary curvature constraint is determined as: wherein, κ max is the maximum curvature limit.

[0114] The existing curvature constraint is: wherein, Therefore, the existing curvature constraint is non-convex and nonlinear, which is difficult to solve in real time, is easy to fall into a local optimum in a non-convex space, and leads to a long solution time when planning a path, which is difficult to meet the real-time requirements of automatic driving path planning.

[0115] Therefore, the present application determines the and wherein, Then, based on the determined and The preliminary curvature constraint is determined as: wherein, k max is the maximum curvature limit.

[0116] Step S450: solving the preliminary curvature constraint by the primal-dual interior point method to obtain the determined λ k1 and μ k1 .

[0117] Step S460: determining the curvature constraint according to λ k1 and μ k1 :

[0118] wherein, the primal-dual interior point method is a numerical algorithm for solving optimization problems, which is used to simultaneously solve the original problem and the dual problem, and find the solutions of the two by gradually optimizing. The present application solves the preliminary curvature constraint by the primal-dual interior point method to obtain the determined λ k1 and μ k1 . Then, the linear curvature constraint is obtained according to λ k1 and μ k1 .

[0119] The present application takes the target function term as a soft constraint, which requires the first derivative of the optimized predicted path point and the reference path point to be as close as possible, and uses the primal-dual interior point method to solve it. Each iteration point will λk and μ k Directly calculated to approximate, by the above method, the nonlinear curvature constraint can be converted into a combination of soft constraint and linear constraint, overcoming the problem that the nonlinear curvature constraint in the Cartesian coordinate path planning method is difficult to solve in real time, and meeting the real-time requirement of path planning.

[0120] Step S470: determining a plurality of trajectory points on the second path as reference path points.

[0121] Step S480: inputting the position information of the reference path points into the objective function for solving to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point meets the constraint condition.

[0122] Step S490: determining a target path according to the position information of a plurality of predicted path points.

[0123] Step S4100: controlling vehicle movement based on the target path.

[0124] Among them, step S470 and step S4100 have been described in detail in the foregoing embodiments, which will not be repeated here.

[0125] Please refer to Figure 5 , Figure 5 A method flow chart of a driving method provided by an embodiment of the application is shown, which specifically includes steps S510 to S5100.

[0126] Step S510: obtaining a first path based on the acquired driving data of the vehicle in the Frenet coordinate system.

[0127] Step S520: converting the acquired first path in the Frenet coordinate system into a second path in the Cartesian coordinate system.

[0128] Among them, step S510 and step S520 have been described in detail in the foregoing embodiments, which will not be repeated here.

[0129] It should be noted that the optimization strategy includes an objective function and a constraint condition.

[0130] In the actual driving process of the vehicle, the curvature change of the curve path actually walked by the vehicle in turning is limited, therefore, the curvature change constraint condition needs to be considered when planning the path.

[0131] The constraint condition of the application further includes a curvature change constraint, which is used to constrain the predicted path point obtained by the objective function to meet the curvature change constraint range, so as to make the accuracy of the obtained target path higher. The detailed content of the objective function is described in the foregoing embodiments, which will not be repeated here.

[0132] Step S530: obtaining the reference curvature change constraint based on the reference path point and wherein,

[0133] Step S540: determining the preliminary curvature change constraint based on the reference curvature change constraint and The preliminary curvature change constraint is determined as follows: wherein, Δκ max is the maximum curvature change limit.

[0134] The existing curvature change constraint is as follows: wherein, Therefore, the existing curvature change constraint is non-convex and nonlinear, which is difficult to solve in real time, is easy to fall into a local optimum in a non-convex space, and leads to a long solving time when planning a path, which is difficult to meet the real-time requirements of automatic driving path planning.

[0135] Therefore, the present application determines the reference curvature change constraint based on the information obtained from the reference path point and wherein, Then, the preliminary curvature change constraint is determined based on the determined reference curvature change constraint and The preliminary curvature change constraint is determined as follows:

[0136] Step S550: solving the preliminary curvature change constraint by the primal-dual interior point method to obtain the determined λ k1 , μ k1 , λ k1+1 and μ k1+1 .

[0137] Step S560: determining the curvature change constraint according to λ k1 , μ k1 , λ k1+1 and μ k1+1 .

[0138] wherein, the preliminary curvature change constraint is solved by the primal-dual interior point method to obtain the determined λ k1 , μ k1 , λ k1+1 and μ k1+1 . Then, the linear curvature change constraint is obtained according to the determined λ k1 , μ k1 , λ k1+1 and μ k1+1 .

[0139] The application takes the objective function The term is taken as a soft constraint, requiring the first derivative of the optimized prediction path point and the reference path point to be as close as possible, and the original dual interior point method is used for solving, and each iteration point will And The direct calculation is approximated, and by the above method, the nonlinear curvature change constraint can be converted into a combination of soft constraints and linear constraints, overcoming the problem that the nonlinear curvature change constraint in the Cartesian coordinate system path planning method is difficult to solve in real time, and meeting the real-time requirements of path planning.

[0140] Step S570: determining a plurality of trajectory points on the second path as reference path points;

[0141] Step S580: inputting the position information of the reference path points into the objective function for solving to obtain a prediction path point corresponding to each reference path point, wherein the prediction path point satisfies the constraint condition;

[0142] Step S590: determining a target path according to the position information of a plurality of the prediction path points.

[0143] Step S5100: controlling vehicle movement based on the target path.

[0144] Among them, step S570 and step S5100 have been described in detail in the foregoing embodiments, which will not be repeated here.

[0145] Please refer to Figure 6 , Figure 6 A method flowchart of a driving method provided by an embodiment of the application is shown, and the method specifically includes steps S610 to S6100.

[0146] Step S610: obtaining a first path based on the acquired driving data of the vehicle in the Frenet coordinate system.

[0147] Step S620: converting the acquired first path in the Frenet coordinate system into a second path in the Cartesian coordinate system.

[0148] Step S630: determining a plurality of trajectory points on the second path as reference path points.

[0149] Among them, step S610 and step S630 have been described in detail in the foregoing embodiments, which will not be repeated here.

[0150] It should be noted that the optimization strategy includes an objective function and a constraint condition.

[0151] The constraints include obstacle constraints, which are used to constrain the predicted path points obtained by the objective function to meet the constraint range of the preset location region, thereby making the accuracy of the obtained target path higher. For details of the objective function, please refer to the foregoing embodiments, which will not be repeated here.

[0152] Step S640: Obtain multiple first trajectory points based on the global path, where the global path is the planned path of the vehicle from the starting point to the end point.

[0153] The global path is the planned path of the vehicle from the starting point to the end point. In other words, the global path is a curve from the starting point to the end point in the Cartesian coordinate system. There are multiple ways to obtain multiple first trajectory points from the global path. The first trajectory points can be obtained based on the curve length or by the curvature. The specific acquisition method is similar to the determination of multiple trajectory points on the second path in the previous embodiment, and will not be described in detail here.

[0154] Step S650: Based on the acquired obstacle position information, determine the boundary point of the normal direction of the first trajectory point, wherein the normal direction is perpendicular to the reference line where the first trajectory point is located.

[0155] Specifically, the location information of obstacles is obtained from the acquired vehicle driving data, and the boundary point of the normal direction of each first trajectory point is determined. The boundary point can be the intersection of the line in the normal direction and the obstacle or road boundary line, or it can be a point at a fixed distance from the obstacle or road boundary line. The normal direction is perpendicular to the reference line where the first trajectory point is located.

[0156] Step S660: Connect the boundary points on the same side of two adjacent first trajectory points, and connect the left and right boundary points of the first trajectory points to obtain multiple convex polygons.

[0157] in, Figure 9 A schematic diagram of a convex polygon provided in an embodiment of this application is shown, such as... Figure 9 As shown, on road 111, the same-side boundary points of two adjacent first trajectory points are connected, and the left and right boundary points of the first trajectory points are connected to obtain multiple convex polygons 112, wherein the obstacle 113 is located outside the area where the convex polygons 112 are located.

[0158] Step S670: Combine the multiple convex polygons into a preset position region.

[0159] in, Figure 10 This application provides a schematic diagram of a preset location area, as shown in the embodiment. Figure 10 As shown, on road 111, multiple convex polygons are combined into a preset position region 211, wherein obstacle 113 is located outside the preset position region 211.

[0160] Step S680: establishing an obstacle constraint based on the preset position region.

[0161] In the Cartesian coordinate system, the obstacle constraint is established based on the preset position region, so that the obstacle constraint represents the preset position region, and the vehicle can only travel in the preset position region. If the vehicle travels outside the preset position region, a traffic accident may occur.

[0162] The existing obstacle constraint is generally a non-convex quadratic constraint, which cannot be quickly solved and is difficult to meet the real-time requirements of automatic driving path planning.

[0163] Therefore, the application proposes a linear obstacle constraint. A convex polygon is constructed by a first trajectory point, and a plurality of convex polygons are merged into a maximum convex hull, that is, a preset position region. The preset position region can be understood as a drivable corridor. The obstacle constraint can be represented as a linear constraint by the drivable corridor, overcoming the problem that the non-linear obstacle constraint in the traditional Cartesian coordinate system path planning method is difficult to solve in real time.

[0164] An exemplary obstacle constraint is: for the drivable corridor.

[0165] Step S690: inputting the position information of the reference path point into the objective function for solving to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies the constraint condition.

[0166] Step S6100: determining a target path according to the position information of a plurality of predicted path points.

[0167] Step S6110: controlling vehicle movement based on the target path.

[0168] Among them, step S690 and step S6110 have been described in detail in the foregoing embodiments, and will not be repeated here.

[0169] In an optional embodiment, the constraint condition includes a system constraint, and the system constraint is: wherein, wherein, s is the arc length between two adjacent path points, and the system constraint is a linear system. The application uses differential flatness to overcome the nonlinearity of the vehicle kinematic model. When solving the objective function, the position information of the reference path point is input into the objective function for solving to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies the system constraint condition.

[0170] In an optional embodiment, the constraint conditions include system constraints, obstacle constraints, curvature constraints, and curvature rate constraints. Details of the system constraints, obstacle constraints, curvature constraints, and curvature rate constraints can be found in the foregoing embodiments, which will not be repeated here.

[0171] Please refer to Figure 7 , Figure 7 A method flowchart of a driving method provided by an embodiment of the application is shown, which specifically includes steps S710 to S760.

[0172] Step S710: In the Frenet coordinate system, a first path is obtained based on the acquired driving data of the vehicle.

[0173] Step S720: The first path in the Frenet coordinate system is converted into a second path in the Cartesian coordinate system.

[0174] Details of steps S710 and S720 have been described in the foregoing embodiments, which will not be repeated here.

[0175] Step S730: It is determined whether the second path meets a preset condition.

[0176] In a large-curvature curve, the local planning in the Frenet coordinate system may not meet the actual kinematic requirements of the vehicle, but in a relatively straight road, the local planning in the Frenet coordinate system meets the actual kinematic requirements of the vehicle, and the vehicle can be controlled based on the local planning path in the Frenet coordinate system. Therefore, it is necessary to determine whether the first path meets the preset condition.

[0177] In an optional embodiment, the preset condition can be a curvature threshold.

[0178] For example, when the curvature of the second path is greater than or equal to the curvature threshold, it indicates that the second path does not meet the preset condition, and when the curvature of the second path is less than the curvature threshold, it indicates that the second path meets the preset condition.

[0179] In an optional embodiment, the preset condition can be a curvature rate threshold.

[0180] For example, when the curvature rate of the second path is greater than or equal to the curvature rate threshold, it indicates that the second path does not meet the preset condition, and when the curvature rate of the second path is less than the curvature rate threshold, it indicates that the second path meets the preset condition.

[0181] Step S740: If the second path does not meet the preset condition, the second path is optimized according to an optimization strategy based on the Cartesian coordinate system to obtain a target path.

[0182] If the second path does not satisfy the preset condition, it indicates that the second path is not the optimal planning path, and the second path needs to be optimized. An optimization strategy is determined based on a Cartesian coordinate system.

[0183] Step S750: If the second path satisfies the preset condition, vehicle movement is controlled based on the second path.

[0184] If the second path satisfies the preset condition, it indicates that the second path is the optimal planning path, and vehicle movement can be controlled based on the second path.

[0185] Step S760: Vehicle movement is controlled based on the target path.

[0186] Step S760 has been described in detail in the foregoing embodiments, and will not be described here.

[0187] Please refer to Figure 8 , Figure 8 A method flowchart of a driving method provided in an embodiment of the application is shown, and the method specifically includes steps S810 to S880.

[0188] Step S810: Global path input.

[0189] The details of the global path are described in the foregoing embodiments, and will not be described here.

[0190] Step S820: Perception information input.

[0191] The perception information is the driving data in the foregoing embodiments.

[0192] Step S830: First path in the Frenet coordinate system is obtained.

[0193] The first path in the Frenet coordinate system is obtained by using the obtained global path information and the perception information.

[0194] Step S840: Driveable corridor is constructed.

[0195] The driveable corridor is constructed by using the obtained global path information and the perception information, and the driveable corridor represents a region in which the vehicle can drive.

[0196] Step S850: Optimization strategy is determined.

[0197] The optimization strategy is determined according to the first path in the Frenet coordinate system and the driveable corridor. It needs to be noted that the first path needs to be converted into a second path in the Cartesian coordinate system, and then the optimization strategy is determined according to the second path. The driveable corridor is used to determine the obstacle constraint.

[0198] Step S860: solving the optimization strategy to obtain the target path.

[0199] For details, please refer to the description in step S130 of the foregoing embodiment, which will not be repeated here.

[0200] Step S870: splicing the target path and the third path.

[0201] After converting the third path in the Frenet coordinate system into the fourth path in the Cartesian coordinate system, the target path and the fourth path are spliced.

[0202] Step S880: obtaining the final planning path.

[0203] Controlling the vehicle according to the final planning path.

[0204] For details, please refer to Figure 11 , Figure 11 A schematic diagram of the target path provided by the embodiment of the application is shown, in which the vehicle 314 travels in the road 111, the first path 311 in the Frenet coordinate system is obtained based on the pre-acquired global path, the first path is converted into the second path in the Cartesian coordinate system by the method of the application, and the target path 312 is obtained by optimizing the second path, and then the target path 312 and the third path 313 in the Frenet coordinate system are spliced, and the splicing point is indicated by “*” in Figure 11 The upper solid line connected with “*” is the target path 312, and the lower solid line connected with “*” is the third path 313.

[0205] It should be noted that after the third path 313 is converted into the fourth path in the Cartesian coordinate system, the target path 312 and the fourth path are spliced. The application combines the short target path with the long fourth path as the final planning path, and through the combination of long and short planning, the dependence on the smoothness of the reference line is reduced, and the predictability and comfort of vehicle travel are improved.

[0206] For details, please refer to Figure 12 , Figure 12 A structural block diagram of a driving device 600 provided by the embodiment of the application is shown, which comprises a conversion unit 610, a construction unit 620, a determination unit 630, and a control unit 640.

[0207] The conversion unit 610 is configured to obtain a first path in a Frenet coordinate system based on acquired driving data of a vehicle.

[0208] The construction unit 620 is configured to convert the acquired first path in the Frenet coordinate system into a second path in a Cartesian coordinate system.

[0209] The determination unit 630 is configured to optimize the second path according to an optimization strategy based on a Cartesian coordinate system to obtain a target path.

[0210] Optionally, for a possible implementation, the determination unit 630 is further configured to obtain a reference path point based on the reference path. and wherein, based on and determine a preliminary curvature constraint as: wherein k max is a maximum curvature limit; the preliminary curvature constraint is solved by a primal-dual interior point method to obtain determined λ k1 and μ k1 ; a curvature constraint is determined according to λ k1 and μ k1 .

[0211] Optionally, for a possible implementation, the determination unit 630 is further configured to obtain a reference path point based on the reference path. and wherein, based on and determine a preliminary curvature change constraint as: wherein Δκ max is a maximum curvature change limit; the preliminary curvature change constraint is solved by a primal-dual interior point method to obtain determined λ k1 , μ k1 , λ k1+1 and μ k1+1 ; a curvature change constraint is determined according to λ k1 , μ k1 , λ k1+1 and μ k1+1 .

[0212] Optionally, for a possible implementation, the determination unit 630 is further configured to obtain a plurality of first trajectory points based on a global path, the global path being a planned path of a vehicle from a starting point to an ending point; determine boundary points of a normal direction of the first trajectory points according to the obtained position information of the obstacles, the normal direction being perpendicular to a reference line on which the first trajectory points are located; connect the boundary points on the same side of two adjacent first trajectory points, and connect the left and right boundary points of the first trajectory points to obtain a plurality of convex polygons; combine the plurality of convex polygons into a preset position region; and establish an obstacle constraint based on the preset position region.

[0213] Optionally, for one possible implementation, the determining unit 630 is further configured to determine whether the second path meets a preset condition; if the second path does not meet the preset condition, the second path is optimized according to an optimization strategy based on a Cartesian coordinate system to obtain a target path.

[0214] Optionally, for one possible implementation, the determining unit 630 is further configured to determine a plurality of trajectory points on the second path as reference path points; input position information of the reference path points into a target function to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point meets the constraint condition; and determine a target path according to position information of the plurality of predicted path points.

[0215] The control unit 640 is configured to control vehicle movement based on the target path.

[0216] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0217] In several embodiments provided in the present application, the coupling between units can be electrical, mechanical or other forms of coupling. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0218] Please refer to Figure 13 , Figure 13 A structural block diagram of an electronic device 700 provided by an embodiment of the present application is shown. The electronic device 700 can be a car machine system, which can be arranged in a vehicle. The electronic device 700 in the present application can include one or more of the following components: a processor 711, a memory 712, and one or more application programs, wherein the processor 711 is electrically connected to the memory 712, and the one or more programs are configured to execute the method described in the foregoing method embodiments.

[0219] The processor 711 can include one or more processing cores. The processor 711 connects various parts within the entire electronic device 700 with various interfaces and lines, performs various functions of the electronic device 700 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 712, and calling data stored in the memory 712. Optionally, the processor 711 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 711 can integrate a combination of one or several of a central processing unit (CPU), a graphics processor (GPU), and a modem. Among them, the CPU is mainly responsible for processing an operating system, a user interface, and a computer program, etc.; the GPU is responsible for rendering and drawing display content; and the modem is responsible for processing wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 711, but be implemented by a communication chip alone. Specifically, the one or more processors 711 can execute the methods described in the foregoing embodiments.

[0220] For some embodiments, the memory 712 can include a random access memory (RAM) and can also include a read-only memory (ROM). The memory 712 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 712 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing each of the methods described below, etc. The data storage area can also store data created by the electronic device 700 in use, etc.

[0221] Please refer to Figure 14 which shows a structural block diagram of a computer readable storage medium provided by an embodiment of the present application. The computer readable medium 800 stores program codes therein, and the program codes can be called and executed by a processor to perform the methods described in the above method embodiments.

[0222] The computer-readable storage medium 800 can be an electronic storage memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk or a ROM. Optionally, the computer-readable storage medium 800 comprises a non-transitory computer-readable medium. The computer-readable storage medium 800 has a storage space for program codes 810 to execute any of the method steps described above. These program codes can be read from or written to one or more computer program products. The program codes 810 can be compressed, for example, in an appropriate form.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent replacements for some of the technical features, without departing from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of driving a vehicle, characterized by, The method comprises: obtaining a first path based on the acquired driving data of the vehicle in a Frenet coordinate system; converting the acquired first path in the Frenet coordinate system into a second path in a Cartesian coordinate system; determining a plurality of trajectory points on the second path as reference path points; inputting the position information of the reference path points into a target function included in an optimization strategy of the Cartesian coordinate system for solving, to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies a constraint condition; the optimization strategy comprises the constraint condition; determining a target path according to the position information of a plurality of predicted path points; controlling vehicle movement based on the target path; wherein the target function is: wherein with denotes a state variable of the objective function, , , denotes a control variable of the objective function, ; wherein , respectively denote the horizontal and vertical coordinates of the predicted path point, , , respectively denote the first, second and third derivative of the horizontal coordinate of the predicted path point with respect to the arc length, denotes the third derivative of the horizontal coordinate of the previous predicted path point, , , respectively denote the first, second and third derivative of the vertical coordinate of the predicted path point with respect to the arc length, denotes the third derivative of the vertical coordinate of the previous predicted path point; wherein , respectively denote the horizontal and vertical coordinates of the reference path point, denotes the first derivative of the horizontal coordinate of the reference path point with respect to the arc length, denotes the first derivative of the vertical coordinate of the reference path point with respect to the arc length, denotes the curvature of the predicted path point, denotes the curvature of the reference path point; wherein, , , , , , is a preset weight, N is the number of prediction path points, and k is a serial number.

2. The method of claim 1, wherein, the constraint condition comprises a curvature constraint, and before the position information of the reference path points is input into the target function for solving to obtain a predicted path point corresponding to each reference path point, the method further comprises: based on the reference waypoint and wherein , ; Based on and The preliminary curvature constraint is determined as: wherein, is the maximum curvature limit; Solving the preliminary curvature constraint by a primal-dual interior point method, a certain and ; According to and The curvature constraint is determined as: .

3. The method of claim 1, wherein, the constraint condition comprises a curvature change constraint, and before the position information of the reference path points is input into the target function for solving to obtain a predicted path point corresponding to each reference path point, the method further comprises: based on the reference waypoint , , and wherein , , , ; Based on , , and determining the preliminary curvature variation constraint is: wherein, is the maximum curvature change limit; solving the preliminary curvature variation constraint by a primal-dual interior point method, obtaining a certain , , and ; According to , , and the curvature variation constraint is determined as: 。 4. The method of claim 1, wherein, the constraint condition comprises an obstacle constraint, and before the position information of the reference path points is input into the target function for solving to obtain a predicted path point corresponding to each reference path point, the method further comprises: obtaining a plurality of first trajectory points based on a global path, wherein the global path is a planned path of the vehicle from a starting point to an ending point; determining a boundary point in the normal direction of the first trajectory point according to the position information of the acquired obstacle, wherein the normal direction is perpendicular to a reference line on which the first trajectory point is located; connecting the same side boundary points of adjacent two first trajectory points and connecting the left and right boundary points of the first trajectory point to obtain a plurality of convex polygons; combining a plurality of the convex polygons into a preset position region; establishing an obstacle constraint based on the preset position region.

5. The method of claim 1, wherein, The determination of a plurality of trajectory points on the second path as reference path points comprises: judging whether the second path satisfies a preset condition; if the second path does not satisfy the preset condition, determining a plurality of trajectory points on the second path as reference path points.

6. A driving apparatus characterized by comprising: The device comprises: a conversion unit configured to obtain a first path based on the acquired driving data of the vehicle in a Frenet coordinate system; a construction unit configured to convert the acquired first path in the Frenet coordinate system into a second path in a Cartesian coordinate system; a determination unit configured to determine a plurality of trajectory points on the second path as reference path points; input the position information of the reference path points into a target function included in an optimization strategy of the Cartesian coordinate system for solving, to obtain a predicted path point corresponding to each reference path point, wherein the predicted path point satisfies a constraint condition; the optimization strategy comprises the constraint condition; determine a target path according to the position information of a plurality of predicted path points; a control unit configured to control vehicle movement based on the target path; wherein the target function is: wherein with denotes a state variable of the objective function, , , denotes a control variable of the objective function, ; wherein , respectively denote the horizontal and vertical coordinates of the predicted path point, , , respectively denote the first, second and third derivative of the horizontal coordinate of the predicted path point with respect to the arc length, denotes the third derivative of the horizontal coordinate of the previous predicted path point, , , respectively denote the first, second and third derivative of the vertical coordinate of the predicted path point with respect to the arc length, denotes the third derivative of the vertical coordinate of the previous predicted path point; wherein , respectively denote the horizontal and vertical coordinates of the reference path point, denotes the first derivative of the horizontal coordinate of the reference path point with respect to the arc length, denotes the first derivative of the vertical coordinate of the reference path point with respect to the arc length, denotes the curvature of the predicted path point, denotes the curvature of the reference path point; wherein, , , , , , is a preset weight, N is the number of prediction path points, and k is a serial number.

7. An electronic device, comprising: comprises: one or more processors; a memory; one or more application programs, wherein the one or more application programs are stored in the memory, the one or more application programs are configured to be executed by the one or more processors, and the one or more application programs are configured to perform the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores program codes, and the program codes can be invoked by the processor to perform the method according to any one of claims 1-5.

Images