Automated parking method, parking control apparatus, and medium and device

Abstract

Disclosed in the present application are an automated parking method, a parking control apparatus, and a medium and a device. The method comprises: performing position sampling on the basis of the start position of a vehicle and a set of historical sampling positions, so as to obtain several current sampling positions; on the basis of each current sampling position and a target position, determining a target cost value corresponding to each current sampling position; determining a target sampling position on the basis of the target cost value for each current sampling position, and adding the target sampling position to the set of historical sampling positions, so as to obtain an updated set of historical sampling positions; determining whether a predetermined position sampling stop condition is met; when the position sampling stop condition is not met, performing the next round of position sampling on the basis of the start position and the updated set of historical sampling positions until the sampling stop condition is met, and generating a parking path on the basis of the set of historical sampling positions; and on the basis of the parking path, controlling the vehicle. The present application can solve the problem of inaccurate automated parking.

Description

An autonomous parking method, parking control device, medium and equipment

[0001] The present application claims priority to the Chinese patent application No. 202410724758.3, filed on June 5, 2024, and entitled "An autonomous parking method, parking control device, medium and equipment", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of intelligent driving, in particular to an autonomous parking method, parking control device, medium and equipment. BACKGROUND

[0003] Autonomous valet parking is an effective means to solve the last mile automatic driving problem, and planning a feasible path based on a parking lot environment map and controlling the vehicle to accurately track the feasible path are core technologies of autonomous valet parking systems, which have been widely focused by OEMs and autonomous driving companies.

[0004] However, the current autonomous parking method has problems such as inaccurate path planning and inaccurate path control, which leads to inaccurate autonomous parking.

[0005] SUMMARY

[0006] Therefore, the present application provides an autonomous parking method, parking control device, medium and equipment, which mainly aims to solve the problem of inaccurate autonomous parking.

[0007] To solve the above problems, the present application provides an autonomous parking method, comprising:

[0008] Sampling positions based on a starting position of a vehicle and a set of historical sampling positions to obtain a plurality of current sampling positions;

[0009] Determining target cost values corresponding to each of the current sampling positions based on each of the current sampling positions and a target position;

[0010] Determining a target sampling position based on the target cost values of each of the current sampling positions, and adding the target sampling position to the set of historical sampling positions to obtain an updated set of historical sampling positions;

[0011] Determining whether a predetermined position sampling stop condition is met; when the position sampling stop condition is not met, performing the next round of position sampling based on the starting position and the updated set of historical sampling positions, and when the position sampling stop condition is met, generating a parking path based on the updated set of historical sampling positions;

[0012] controlling the vehicle to travel along the parking path based on the parking path.

[0013] The autonomous parking method in the application can make the determination of the current sampling position more reasonable by randomly sampling in real time according to the starting position and the historical sampling position set, can filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate, and the cycle is repeated until the position sampling is completed, that is, the target sampling position obtained in each round of sampling (that is, based on each historical sampling position in the historical sampling set) is used to generate the parking path, so that the generation of the parking path is more reasonable and accurate, and the autonomous parking can be accurately performed based on the parking path.

[0014] Optionally, the position sampling based on the starting position of the vehicle and the historical sampling position set to obtain a plurality of current sampling positions specifically includes:

[0015] Based on the starting position of the vehicle and each historical sampling position in the historical sampling position set, a current sampling region and a plurality of expansion sampling regions adjacent to the current sampling region are determined.

[0016] Based on the sampling probability corresponding to each sampling region, the current sampling region and each expansion sampling region are respectively sampled to obtain the plurality of current sampling positions.

[0017] In the application, the current sampling region is first determined, and then the expansion sampling region is determined according to the current sampling region, and the position sampling can be reasonably and accurately sampled based on the current sampling region and the expansion sampling region, so that the plurality of current sampling positions obtained are more reasonable and accurate, thereby laying a foundation for reasonably and accurately generating the parking path.

[0018] Optionally, before the random sampling, the method further includes determining the sampling probability corresponding to the current sampling region and each expansion sampling region, specifically including:

[0019] A first sampling probability corresponding to the current sampling region is determined.

[0020] Based on the area of each expansion sampling region and the first sampling probability, a second sampling probability corresponding to each expansion sampling region is determined.

[0021] In the present application, by further determining the sampling probabilities corresponding to different sampling regions, the subsequent position sampling on the current sampling region and the expanded sampling region can be made more reasonable and accurate, thereby further ensuring the rationality and accuracy of the current sampling position, and laying a foundation for subsequent parking path planning based on each current sampling position.

[0022] Optionally, the target cost values corresponding to each of the current sampling positions are determined based on each of the current sampling positions and the target position, and specifically include:

[0023] A target state corresponding to the target position is determined based on the target position.

[0024] Based on the target state and a plurality of candidate arrival times corresponding to each of the current sampling positions, initial cost values corresponding to each of the candidate arrival times are calculated and obtained.

[0025] Based on each initial cost value corresponding to the same current sampling position, the minimum initial cost value is determined as the target cost value of the current sampling position, so as to obtain the target cost values corresponding to each of the current sampling positions.

[0026] In the present application, by determining the target state corresponding to the target position, the initial cost values of each candidate arrival time can be reasonably and accurately calculated based on the target state and each candidate arrival time of each current sampling position, thereby laying a foundation for accurately determining the target cost value of each current sampling position based on the initial cost values of each current sampling position.

[0027] Optionally, the target sampling position is determined based on the target cost values of each of the current sampling positions, and specifically includes:

[0028] The current sampling position corresponding to the minimum target cost value is determined as the target current sampling position based on the target cost values of each of the current sampling positions.

[0029] In the present application, by determining the target current sampling position from each of the current sampling positions based on the target cost values of each of the current sampling positions, a foundation is laid for reasonably and accurately generating a parking path based on the target current sampling position.

[0030] Optionally, the determination of whether the predetermined position sampling stop condition is met includes:

[0031] Whether the predetermined position sampling stop condition is met is judged based on the distance between the target sampling position and the target position and / or the number of sampling rounds.

[0032] In the application, the sampling stop judgment is based on the distance and the sampling wheel number, so that the sampling stop judgment is more accurate, and the problems of over-sampling or under-sampling are avoided.

[0033] Optionally, the vehicle is controlled to travel along the parking path based on the parking path, specifically comprising:

[0034] A next travel position is determined from the parking path based on a current travel position in real time;

[0035] Based on the current travel position and the next travel position, a lateral error and a longitudinal error of the vehicle are calculated;

[0036] Based on the lateral error and the longitudinal error, a yaw rate of the vehicle is calculated;

[0037] Based on the yaw rate, a front wheel steering angle of the vehicle is calculated;

[0038] The vehicle is controlled based on the front wheel steering angle, so that the vehicle travels along the parking path.

[0039] In the application, the next travel position is determined based on the parking path, so that the lateral error and the longitudinal error of the vehicle are reasonably and accurately determined based on the current position and the next travel position, and then the travel parameters of the vehicle, i.e., the yaw rate and the front wheel steering angle of the vehicle, are accurately determined based on the lateral error and the longitudinal error of the vehicle, so that the autonomous parking can be accurately performed based on the travel parameters.

[0040] To solve the above problems, the application provides a parking control device, comprising:

[0041] A sampling module is configured to perform position sampling based on a starting position of a vehicle and a set of historical sampling positions, to obtain a plurality of current sampling positions;

[0042] A first determination module is configured to determine a target cost value corresponding to each of the current sampling positions based on each of the current sampling positions and a target position, respectively;

[0043] A second determination module is configured to determine a target sampling position based on the target cost value of each of the current sampling positions, and add the target sampling position to the set of historical sampling positions to obtain an updated set of historical sampling positions;

[0044] A generation module is configured to determine whether a predetermined position sampling stop condition is met, perform next round of position sampling based on the starting position and the updated set of historical sampling positions when the position sampling stop condition is not met, and generate a parking path based on the updated set of historical sampling positions when the position sampling stop condition is met.

[0045] a control module configured to control the vehicle to travel along the parking path based on the parking path.

[0046] The device in the application can determine the current sampling positions more reasonably by randomly sampling in real time according to the starting position and the historical sampling position set, can filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate, and the cycle is repeated until the position sampling is completed, that is, the target sampling position obtained in each round of sampling (that is, based on each historical sampling position in the historical sampling set) is used to produce the parking path, so that the generation of the parking path is more reasonable and accurate, and the subsequent autonomous parking can be accurately based on the parking path.

[0047] To solve the above problems, the application provides a storage medium, the storage medium stores a computer program, and the computer program is executed by a processor to realize the steps of the autonomous parking method.

[0048] The storage medium in the application can determine the current sampling positions more reasonably by randomly sampling in real time according to the starting position and the historical sampling position set, can filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate, and the cycle is repeated until the position sampling is completed, that is, the target sampling position obtained in each round of sampling (that is, based on each historical sampling position in the historical sampling set) is used to produce the parking path, so that the generation of the parking path is more reasonable and accurate, and the subsequent autonomous parking can be accurately based on the parking path.

[0049] To solve the above problems, the application provides an electronic device, at least comprising a memory and a processor, the memory stores a computer program, and the processor realizes the steps of the autonomous parking method when executing the computer program on the memory.

[0050] The electronic device in the application can determine the current sampling positions more reasonably by randomly sampling according to the starting position and the historical sampling position set, and can filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate, and the cycle is repeated until the position sampling is completed, that is, the target sampling position obtained in each round of sampling (that is, based on the historical sampling positions in the historical sampling set) is used to generate the parking path, so that the generation of the parking path is more reasonable and accurate, and subsequent autonomous parking can be accurately performed based on the parking path.

[0051] The above description is only a summary of the technical solutions of the application. In order to enable the technical means of the application to be more clearly understood, the application can be implemented according to the content of the specification, and in order to enable the above and other purposes, characteristics and advantages of the application to be more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0052] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the scope of the present application. Furthermore, the same reference numerals are used throughout the several views to denote the same or similar parts. In the drawings:

[0053] Fig. 1 is a flowchart of an autonomous parking method according to an embodiment of the application;

[0054] Fig. 2 is a schematic diagram of the position relationship of a sampling area according to an embodiment of the application;

[0055] Fig. 3 is a schematic diagram of a parking path tracking error according to an embodiment of the application;

[0056] Fig. 4 is a structural block diagram of a parking control device according to another embodiment of the application;

[0057] Fig. 5 is a structural block diagram of an electronic device according to another embodiment of the application. DETAILED DESCRIPTION

[0058] The various solutions and features of the application are described herein with reference to the accompanying drawings.

[0059] It should be understood that various modifications can be made to the embodiments of the application described herein. Therefore, the above description should not be regarded as limiting, but only as an example of the embodiments. Those skilled in the art will think of other modifications within the scope and spirit of the application.

[0060] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and, together with the description given above and the detailed description of the embodiments given below, serve to explain the principles of the application.

[0061] These and other characteristics of the present application will become apparent from the following description and the associated drawings, wherein the preferred forms of the application are given, by way of non-limiting examples.

[0062] It should also be understood that, although the present application has been described above with reference to particular means, materials and embodiments, the application is not intended to be limited to the particulars described. Rather, the application extends to all functionally equivalent structures, methods and uses.

[0063] The above and other aspects, features and advantages of the present application will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings, when considered in conjunction with the following detailed description.

[0064] Specific embodiments of the present application are described hereinafter, with reference to the accompanying drawings; however, it will be understood that the application is not limited to the embodiments described, but can be carried out in various ways. Well-known and / or redundant functions and structures are not described in detail to avoid obscuring the application unnecessarily. Therefore, specific structural and functional details disclosed herein are not intended to limit the application, but merely as a basis for the claims and a representative basis for teaching one skilled in the art to variously employ the present application in virtually any appropriate detail.

[0065] The specification can use phrases such as "in one embodiment", "in another embodiment", "in yet another embodiment", or "in other embodiments", which can refer to one or more of the same or different embodiments of the application.

[0066] The embodiments of the present application provide an autonomous parking method, as shown in FIG. 1, the method in the embodiments includes the following steps:

[0067] In step S101, position sampling is performed based on a starting position of a vehicle and a historical sampling position set to obtain a plurality of current sampling positions.

[0068] In this step, the historical sampling position set contains historical sampling positions obtained in previous rounds of position sampling. If it is the first time of position sampling, the historical sampling position set is empty, i.e., does not contain historical sampling positions, so that position sampling can be directly performed according to the starting position.

[0069] In this step, when performing position sampling, a plurality of sampling regions can be determined according to the starting position or according to the starting position and each historical sampling position in the historical sampling set, and sampling probabilities corresponding to the sampling regions are determined, and then position sampling can be performed on the corresponding sampling regions according to the sampling probabilities, so as to obtain a plurality of current sampling positions.

[0070] In step S102, target cost values corresponding to the current sampling positions are determined based on the current sampling positions and the target position respectively.

[0071] In this step, the target position refers to the end position to which the vehicle is to travel. After obtaining a plurality of current sampling positions, for each sampling position, the target state of the target position and an arbitrary arrival time are used to calculate a cost value corresponding to the arrival time, and then the minimum cost value is determined from the cost values as the target cost value of the current sampling position. That is, for any one current sampling position, since there can be multiple possible arrival times from the current sampling position to the target position, i.e., different cost values corresponding to different arrival times, the minimum cost value calculated can be taken as the target cost value of the current sampling position, and thus the target cost value of each current sampling position can be determined.

[0072] In step S103, a target sampling position is determined based on the target cost values of the current sampling positions, and the target sampling position is added to the set of historical sampling positions to obtain an updated set of historical sampling positions.

[0073] In this step, after determining the target cost values of the current sampling positions, the current sampling position with the minimum target cost value can be selected from the current sampling positions as the target sampling position for use in subsequent generation of a parking path.

[0074] In step S104, it is determined whether a predetermined position sampling stop condition is met. When the position sampling stop condition is not met, the next round of position sampling is performed based on the starting position and the updated set of historical sampling positions, and when the position sampling stop condition is met, a parking path is generated based on the updated set of historical sampling positions.

[0075] In this step, it can be determined whether the predetermined position sampling stop condition is met according to the target sampling position determined in step S103, and / or whether the predetermined position sampling stop condition is met based on the number of sampling rounds. For example, when the distance between the target sampling position and the target position is less than a predetermined distance threshold, or the number of sampling rounds is greater than a predetermined sampling round threshold, it is determined that the predetermined position sampling stop condition is met; otherwise, the sampling stop condition is not met.

[0076] In step S105, the vehicle is controlled to travel along the parking path based on the parking path.

[0077] In this step, after obtaining the parking path, the front wheel steering angle of the vehicle can be determined according to the current position of the vehicle and the next position in the parking path to be reached, so that the vehicle is controlled based on the front wheel steering angle to drive along the parking path to the target position.

[0078] The autonomous parking method in this embodiment can make the determination of the current sampling position more reasonable by randomly sampling in real time according to the starting position and the set of historical sampling positions, and then can filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate. This cycle is repeated until the position sampling is completed, and the parking path can be generated based on the target sampling position obtained in each round of sampling (i.e., each historical sampling position in the set of historical sampling positions). That is, by using the iterative sampling method to determine the plurality of current sampling positions of the next path position point, and combining the cost calculation to determine the target sampling position as the next path position point from the current sampling positions, the determination of the target sampling position is more reasonable and accurate, and the parking path generated based on the target sampling position is more reasonable and accurate, so that the autonomous parking can be accurately performed based on the parking path.

[0079] On the basis of the above-mentioned embodiments, another embodiment of the present application provides an autonomous parking method. In this embodiment, when the position sampling is performed based on the starting position of the vehicle and the set of historical sampling positions to obtain a plurality of current sampling positions, a plurality of sampling regions can be determined according to the starting position or according to the starting position and each historical sampling position in the set of historical sampling positions, and the sampling probability corresponding to each sampling region is determined. Then, the position sampling can be performed on the corresponding sampling region according to the sampling probability, so as to obtain a plurality of current sampling positions.

[0080] That is, step S101 in the above-mentioned embodiment specifically includes:

[0081] Step S1011, determining a current sampling region and a plurality of extended sampling regions adjacent to the current sampling region based on the starting position of the vehicle and each historical sampling position in the set of historical sampling positions.

[0082] Step S1012, performing position sampling on the corresponding current sampling region and each extended sampling region based on the sampling probability corresponding to each sampling region, to obtain the plurality of current sampling positions.

[0083] Specifically, in the embodiment, when the step S1011 is performed, i.e., when the current sampling region and each extended sampling region are determined, the current sampling region S0 and the four extended sampling regions S1-S4 can be determined as shown in FIG. 2. When the current sampling region S0 is determined, if the historical sampling position set is empty, i.e., it is the first time to perform position sampling, a seed point can be determined according to the starting position, and then a line connecting the starting position and the seed point is taken as a diagonal line of a rectangle, so as to construct a rectangular region, and the rectangular region is taken as the current sampling region S0. When the historical sampling position set is not empty, i.e., it is not the first time to perform position sampling, four edge position points corresponding to four different directions can be determined according to the starting position and each historical sampling position, for example, four edge position points in the “up”, “down”, “left” and “right” directions are determined first, or four edge position points in the “east”, “south”, “west” and “north” directions are determined first; then, a rectangular region for enveloping the starting position and each historical sampling position is constructed based on the edge position points, and the rectangular region is taken as the current sampling region S0. After the current sampling region S0 is determined, four extended sampling regions adjacent to the current sampling region, i.e., the first extended sampling region S1, the second extended sampling region S2, the third extended sampling region S3 and the fourth extended sampling region S4, can be constructed based on the current sampling region S0 and the predetermined region width.

[0084] In the embodiment, after the current sampling region S0, the first extended sampling region S1, the second extended sampling region S2, the third extended sampling region S3 and the fourth extended sampling region S4 are determined, i.e., after the step S1011 is performed, the sampling probabilities corresponding to the sampling regions can be further determined, i.e., the sampling probabilities corresponding to the regions S0-S4 are determined, which can be determined in the following manner:

[0085] Step one, determining a first sampling probability corresponding to the current sampling region.

[0086] Specifically, the predetermined sampling probability can be taken as the first sampling probability of the current sampling region S0, or a sampling probability corresponding to the region area of the current sampling region S0 is determined as the first sampling probability of the current sampling region S0 based on the region area of the current sampling region S0. That is, a fixed sampling probability P can be preset, and the fixed sampling probability P is taken as the first sampling probability of the current sampling region in each position sampling. Alternatively, a mapping relationship between the region area and the sampling probability can be established in advance, and the first sampling probability P corresponding to the current sampling region is dynamically obtained by searching the mapping relationship according to the region area of the current sampling region.

[0087] Step two, based on the area of each extended sampling region and the first sampling probability, determine the second sampling probability corresponding to each extended sampling region.

[0088] Specifically, the total extended area ΔS and the second sampling probability P n may be calculated as follows: ΔS = ΔS1 + ΔS2 + ΔS3 + ΔS4 P n ' = (1-P)ΔS n / ΔS, n = 1, 2, 3, 4

[0089] Wherein, ΔS1, ΔS2, ΔS3 and ΔS4 respectively represent the area of the first extended sampling region S1, the second extended sampling region S2, the third extended sampling region S3, and the fourth extended sampling region S4. P n ' represents the second sampling probability corresponding to the nth extended sampling region; P represents the first sampling probability.

[0090] That is, the second sampling probability P'1 corresponding to the first extended sampling region S1 is: P'1 = (1-P)ΔS1 / ΔS.

[0091] The second sampling probability P'2 corresponding to the second extended sampling region S2 is: P'2 = (1-P)ΔS2 / ΔS.

[0092] The second sampling probability P'3 corresponding to the third extended sampling region S3 is: P'3 = (1-P)ΔS3 / ΔS.

[0093] The second sampling probability P'4 corresponding to the fourth extended sampling region S4 is: P'4 = (1-P)ΔS4 / ΔS.

[0094] In this embodiment, after the first sampling probability corresponding to the current sampling region is determined and the second sampling probability corresponding to each extended sampling region is calculated, the corresponding sampling region can be respectively sampled based on the sampling probability, so that the current sampling position obtained by sampling is more reasonable and accurate, which lays a foundation for subsequent parking path planning based on each current sampling position.

[0095] Another embodiment of the application provides an autonomous parking method. In this embodiment, when determining the target cost value corresponding to each current sampling position based on each current sampling position and the target position, that is, when performing step S102 in the above embodiment, the following method can be used:

[0096] Step S1021, determining a target state corresponding to the target position based on the target position;

[0097] In this step, the target position represents a terminal position to which the vehicle is to travel, and the target position is specifically represented as x goal For example, according to the target position x goal , the corresponding target state ξ1 is determined. The target state ξ1 includes: the longitudinal speed v x of the vehicle, the curvature p, the coordinate (x, y) of the midpoint of the rear axle, and the azimuth angle

[0098] Step S1022, based on a plurality of candidate arrival times corresponding to each of the current sampling positions, and the target state, calculating to obtain an initial cost value corresponding to each of the candidate arrival times;

[0099] In this step, for any one current sampling position, since the arrival time from the current sampling position to the target position is different, the corresponding vehicle control amount will be different, that is, the vehicle control strategy will be different, which in turn leads to the cost value from the current sampling position to the target position being different, so the same current sampling position can calculate a plurality of initial cost values.

[0100] In this step, when calculating the cost value, the following steps can be sampled:

[0101] Step S1022-1, based on the target state ξ1, using the following matrix calculation formula, calculating to obtain a first matrix A and a second matrix B;

[0102] Step S1022-2, based on the first matrix A, the second matrix B, and the predetermined weight R, using the following state calculation formula, calculating to obtain the intermediate parameter G(τ) corresponding to each candidate arrival time τ of any current sampling position, and the current state

[0103] Step S1022-3, based on each candidate arrival time τ, the target state ξ1, and the intermediate parameter G(τ) corresponding to each candidate arrival time τ and the current state of the current sampling position corresponding to each candidate arrival time τ , using the following cost calculation formula, calculating to obtain the initial cost c(τ) of the current sampling position at each candidate arrival time;

[0104] Step S1023, based on each initial cost value corresponding to the same current sampling position, determining the minimum initial cost value c(τ *) is the target cost value of the current sampling position, so as to obtain the target cost value corresponding to each current sampling position.

[0105] In the embodiment, after obtaining the target cost value corresponding to each current sampling position, the current sampling position corresponding to the minimum target cost value can be further determined as the target current sampling position.

[0106] In the embodiment, the specific calculation principle of the minimum cost value c(τ * ) is as follows:

[0107] Firstly, (x, y) and represent the coordinates and azimuth angle of the midpoint of the rear axle of the automobile, and the kinematic model describing the low-speed driving state of the automobile can be represented as:

[0108] In the formula, v x represents the longitudinal speed of the automobile, p represents the curvature, and (x, y) represents the coordinates of the midpoint of the rear axle. represents the azimuth angle.

[0109] The state quantity of the system is defined as the control quantity is u = [u v u ρ ] T , and formula (1) can be rewritten as:

[0110] The first-order Taylor expansion of formula (2) at (ξ1, u1) is:

[0111] Define and , formula (3) can be rewritten as:

[0112] In the formula, the matrices A and B can be represented as:

[0113] Considering the energy consumption and timeliness of the local linear subsystem described by formula (4) from the starting state ξ0 to the target state ξ1 (i.e., the target position x goal corresponding to the target state), the following control performance index is defined as the kinematic RRT* algorithm cost criterion.

[0114] Specifically, the optimal control performance index at any arrival time τ can be represented as:

[0115] In the formula, G(τ) and are the solutions of formula (7) and formula (8) at any arrival time τ, respectively.

[0116] The fourth-order Runge-Kutta numerical method is used to solve G(τ) and The minimum value c(τ * ) of the control performance index in the iterative calculation process is recorded. * According to formula (6), the control performance index satisfies c(τ) > τ, and the termination condition of the iterative calculation process is c(τ * ) < τ, and c(τ * ) is taken as the control performance index of the optimal arrival time τ

[0117] In this embodiment, after the minimum cost value c(τ * ) is determined, the optimal control input corresponding to the control performance index of the optimal arrival time τ * may be further calculated, which can be represented as: u(t) = R -1 B T η(t), t ∈ [0, τ * ] (9)

[0118] In formula (10), η(t) is the solution of formula (10) at any time t ∈ [0, τ * ].

[0119] The fourth-order Runge-Kutta numerical method is used to solve ξ(t) and η(t) at any time t ∈ [0, τ * ], and then the optimal control input u(t) and the optimal state quantity ξ(t) corresponding to the control performance index of the optimal arrival time τ * are obtained.

[0120] Another embodiment of the present application provides an autonomous parking method. In this embodiment, whether the predetermined position sampling stop condition is met is determined, which can be based on the number of sampling rounds and / or the distance between the target sampling position and the target position.

[0121] That is, the number of sampling rounds can be compared with the predetermined sampling round threshold value. If the number of sampling rounds is less than or equal to the predetermined sampling round, it means that the predetermined position sampling stop condition is not met. On the contrary, if the number of sampling rounds is greater than the predetermined sampling round, it means that the predetermined position sampling stop condition is met.

[0122] In the embodiment, in the case that the number of sampling rounds is less than or equal to the predetermined number of sampling rounds, a target distance between the target sampling position obtained in the latest round of sampling and the target position can be further calculated, and then the target distance is compared with the predetermined distance threshold; if the target distance is less than or equal to the predetermined distance threshold, it is indicated that the predetermined position sampling stop condition is met; otherwise, if the distance is greater than the predetermined distance threshold, it is indicated that the predetermined position sampling stop condition is not met.

[0123] In the embodiment, the judgment of the position sampling is performed in the above manner, so that the judgment result is more reasonable and accurate, and the rationality of the sampling position is ensured, thereby laying a foundation for subsequent path planning based on the sampling position.

[0124] On the basis of the above embodiment, another embodiment of the application provides an autonomous parking method. In the embodiment, when the vehicle is controlled to travel along the parking path based on the parking path, the front wheel steering angle can be accurately calculated by using the method in the embodiment, and then the vehicle is controlled according to the front wheel steering angle, so that the vehicle can strictly travel along the planned parking path, and the problem that the autonomous parking is not accurate and reliable due to the deviation of the vehicle from the parking path can be avoided. In the embodiment, the calculation process of the front wheel steering angle is as follows.

[0125] In step S1051, the next travel position is determined from the parking path based on the current travel position in real time.

[0126] In step S1052, the lateral error and the longitudinal error of the vehicle are calculated based on the current travel position and the next travel position.

[0127] In step S1053, the yaw angular velocity of the vehicle is calculated based on the lateral error and the longitudinal error.

[0128] In the specific implementation process, the yaw angular velocity can be calculated by using the following formula: x(t)=[x1(t) x2(t)] T (Formula Three) γ(t-τ)=α(x(t-τ))+β(x(t-τ))υ(t-τ) (Formula Six)

[0129] wherein, x(t)=[x1(t) x2(t)] x1(t) represents the lateral error, which is known. x2(t) represents the longitudinal error, which is known. l x1(t) also represents the lateral position deviation of the midpoint of the rear axle of the vehicle from the target point on the target path. x υ(t) represents the speed, which is known. ρ is the yaw angle deviation between the midpoint of the rear axle of the vehicle and the next driving position point (target point) on the parking path; e is the curvature at the next driving position point (target point) on the parking path; l This represents the lateral positional deviation between the midpoint of the rear axle of the vehicle and the target point (next driving position) on the parking path; τ is the lag time. γ represents the intermediate control variable, i.e., the yaw rate.

[0130] Matrix (A) c B c Form a controllable pair. B c =[0 1] T κ1>0, κ2>0. sgn(·) is the sign function. C is a pre-defined parameter; υ(θ) represents the velocity, which is known. and x a The estimates of parameter η and the anti-saturation compensation amount are respectively expressed as follows:

[0131] In other words, according to Lateral error and Linear error, v x The yaw rate γ can be calculated using equations (1), (2), (3), (4), (5), and (6).

[0132] Step S1054: Calculate the front wheel steering angle of the vehicle based on the yaw rate;

[0133] In this step, the front wheel steering angle δ can be calculated based on the yaw rate γ using the following (Equation 7). f .

[0134] Where L is the vehicle wheelbase, v x Represents velocity, which is known.

[0135] Step S1055: Control the vehicle based on the front wheel steering angle so that the vehicle travels according to the parking path.

[0136] In this step, the front wheel steering angle δ is calculated. f Then, the vehicle can be controlled based on the front wheel steering angle to achieve precise path tracking.

[0137] By sampling the calculation method described above in this application, the continuous and smooth calculation output of the vehicle's front wheel steering angle can be guaranteed, solving the problem that the autonomous valet parking path tracking is not accurate and reliable due to the complex nonlinear relationship between the vehicle's yaw motion, longitudinal motion, and chassis actuators.

[0138] In this embodiment, the parking path tracking control principle is as follows:

[0139] Since, in the autonomous valet parking process, the path tracking control can need to calculate the output of the front wheel steering angle of the car according to the deviation of the current driving path of the car and the target path, so that the car can quickly, accurately and stably track the target path. Since the car is in a low-speed driving state during the autonomous valet parking process, the lateral sliding of the car caused by the tire side slip characteristic can be ignored, and then the kinematic equation of the car during the autonomous valet parking process is used to describe the driving state. Therefore, the application proposes to establish an automatic valet parking system path tracking error kinematic equation based on the Serret-Frenet coordinate system, which lays a foundation for the design of subsequent control strategies.

[0140] The autonomous valet parking path tracking error schematic diagram is shown in FIG. 3, {N}, {F} and {B} respectively represent the inertial coordinate system, the Serret-Frenet coordinate system with the nearest point to the midpoint of the rear axle of the car on the target path as the origin and the fixed coordinate system with the midpoint of the rear axle of the car as the origin. It is known that the vector radius of the midpoint of the rear axle of the car relative to the origin of the inertial coordinate system {N} is r B / N , the vector radius of the origin of the Serret-Frenet coordinate system {F} relative to the origin of the inertial coordinate system {N} is r F / N , and the vector radius of the midpoint of the rear axle of the car relative to the origin of the Serret-Frenet coordinate system {F} is r B / F , then r B / N =r F / N +r B / F (11)

[0141] In the Serret-Frenet coordinate system {F}, the absolute derivative of the vector radius r B / N in the Serret-Frenet coordinate system {F} can be obtained by differentiating equation (11), and the absolute derivative of the vector radius r F / N in the Serret-Frenet coordinate system {F} is

[0142] In the formula, and are the absolute derivative of the vector radius r B / F in the Serret-Frenet coordinate system {F}, the relative derivative of the vector radius r B / F in the Serret-Frenet coordinate system {F}, the angular velocity vector of the Serret-Frenet coordinate system {F} in the Serret-Frenet coordinate system {F}, and the vector radius r F / N in the Serret-Frenet coordinate system {F}, respectively.

[0143] where s is the arc length of the target path curve; is the angle between the x-axis of the Serret-Frenet coordinate system {F} and the x-axis of the inertial coordinate system {N}; e l is the distance between the midpoint of the rear axle of the vehicle and the origin of the Serret-Frenet coordinate system. If the origin of the Serret-Frenet coordinate system is chosen as the target point, then e l is also denoted as the lateral position deviation of the midpoint of the rear axle of the vehicle from the target point on the target path.

[0144] If the absolute derivative of the radius vector r B / N in the body-fixed coordinate system {B} is denoted as then and satisfy the following coordinate transformation relationship

[0145] where is the yaw angle deviation of the midpoint of the rear axle of the vehicle from the target point on the target path, which can be expressed as

[0146] where is the yaw angle of the vehicle.

[0147] Substituting equation (12) into equation (17), we obtain

[0148] According to the Frenet formula, we have

[0149] where p is the curvature of the target point on the target path.

[0150] Substituting equation (20) into equation (19), we obtain the path tracking error kinematics equation of the autonomous valet parking system as

[0151] where y is the yaw rate of the vehicle, which can be expressed as

[0152] where d f is the steering angle of the front wheels of the vehicle; L is the wheelbase of the vehicle.

[0153] Combining equation (22), from the constraint |d f |≤d fmax , we can know that

[0154] Considering the time delay phenomenon caused by the inertia of the steering system of the vehicle and the information transmission delay of the on-board network, equation (21) is modified as

[0155] where τ is the time delay.

[0156] The path tracking error kinematic equation of the autonomous valet parking system described in equation (24) is first transformed into an unknown disturbance term of the system by using the feedback linearization method, and then an adaptive anti-windup sliding mode control method is used to design the path tracking control strategy of the autonomous valet parking system, so that the closed-loop system has strong robustness to unknown disturbances under the amplitude constraint of the front wheel steering angle of the vehicle, and ensures the continuity and smoothness of the calculated output of the front wheel steering angle of the vehicle. To this end, the following coordinate transformation is defined

[0157] From equation (25), it can be seen that x(t) = [x1(t) x2(t)] T is norm-bounded, i.e. ||x(t)||≤m1. At the same time, the coordinate transformation defined by equation (25) linearizes equation (24) as

[0158] where A c , B c , β -1 (x(t)) and α(x(t)) can be expressed as B c = [0 1] T (28)

[0159] From equations (27) and (28), it can be seen that the matrix (A c , B c ) forms a controllable pair. The control input of the path tracking error linearization equation (26) of the autonomous valet parking system is defined as γ(t-τ) = α(x(t-τ))+β(x(t-τ))υ(t-τ) (31)

[0160] where υ(t-τ) is the new control input, and from equation (23), it can be seen that υ min ≤υ(t-τ)≤υ max (32)

[0161] where υ min and υ max can be expressed as

[0162] Substituting equation (31) into equation (26), the path tracking error linearization equation of the autonomous valet parking system can be modified as

[0163] where d(t) is an unknown disturbance term satisfying ||d(t)||≤m2||x(t)||, which can be expressed as

[0164] For the path tracking error linearization equation of the autonomous valet parking system described by equation (35), the following coordinate transformation is defined

[0165] From equation (37), if the new control input satisfies then

[0166] Combining ||d(t)||≤m2||x(t)||, equation (38) can be obtained

[0167] where η=m2m3 is an unknown constant.

[0168] Using the coordinate transformation defined by equation (37), equation (35) is transformed into

[0169] From equation (40), the vehicle yaw rate time delay term in equation (24) has been transformed into the unknown disturbance term in equation (40). At the same time, considering the new control input amplitude constraint in equation (40), equation (40) is rewritten as

[0170] where sat(·) is a saturation function, which can be expressed as

[0171] For equation (41), the sliding surface is defined as and based on the sliding surface, the path tracking control law of the autonomous valet parking system is designed so that the sliding surface is reachable, and thus Theorem 1 is obtained.

[0172] For equation (41), the following path tracking control law of the autonomous valet parking system is designed

[0173] where κ1>0; κ2>0; sgn(·) is a sign function; and x a are the estimation of the parameter η and the anti-windup compensation, respectively, which can be expressed as

[0174] where κ3>0; κ4>0; Δu(t)=u(t)-υ(t); is the parameter estimation error, which is updated in real time using the following adaptive law

[0175] If the parameters κ1 and κ2 satisfy

[0176] then the sliding surface is reachable.

[0177] Proof: Choose the Lyapunov function as

[0178] Differentiate equation (49), we have

[0179] Substitute inequality (39) and equation (41) into equation (50), we have

[0180] Substitute equation (43), equation (45) and equation (46) into inequality (51), we have

[0181] By Young's inequality, we have

[0182] Substitute inequalities (53)-(55) into inequality (52), we have

[0183] From inequalities (47), (48) and (56), the sliding surface is reachable. Meanwhile, from equation (43), inequality holds.

[0184] Next, we prove that the state trajectories of the closed-loop system on the sliding surface asymptotically converge to the equilibrium point. To this end, define the coordinate transformation as

[0185] Using the coordinate transformation defined by equation (57), equation (40) is transformed into

[0186] Suppose and C = [C1 C2], then equation (58) can be simplified as

[0187] From equation (59), if z1(t) converges to the equilibrium point, then z2(t) also converges to the equilibrium point. To this end, choose the Lyapunov function as

[0188] Differentiate equation (60), we have

[0189] Substitute inequality (49) into inequality (61), we have

[0190] From equation (57) and equation (59), we have

[0191] Substituting equation (63) into inequality (62), we have

[0192] According to inequality (64), if the parameter C = [C1 C2] satisfies

[0193] then the state trajectory of the closed-loop system on the sliding surface converges to the equilibrium point asymptotically.

[0194] In this embodiment, an autonomous valet parking path tracking control method is proposed. Based on the Serret-Frenet coordinate system, a kinematic equation of the automatic valet parking system path tracking error containing the time delay term of the vehicle yaw rate is established. The feedback linearization method is used to convert the time delay term of the vehicle yaw rate into an unknown disturbance term of the system. An adaptive anti-windup sliding mode control method is used to design the autonomous valet parking system path tracking control strategy. The closed-loop system has strong robustness to unknown disturbance terms under the constraint condition of the amplitude of the front wheel steering angle of the vehicle, and ensures the continuity and smoothness of the calculated output of the front wheel steering angle of the vehicle.

[0195] Another embodiment of the present application provides a parking control device, as shown in FIG. 4, which comprises:

[0196] A sampling module 11 is configured to sample positions based on a starting position of a vehicle and a set of historical sampling positions to obtain a plurality of current sampling positions.

[0197] A first determining module 12 is configured to determine a target cost value corresponding to each of the current sampling positions based on each of the current sampling positions and a target position.

[0198] A second determining module 13 is configured to determine a target sampling position based on the target cost value of each of the current sampling positions, and add the target sampling position to the set of historical sampling positions to obtain an updated set of historical sampling positions.

[0199] A generating module 14 is configured to determine whether a predetermined position sampling stop condition is met. When the position sampling stop condition is not met, the next round of position sampling is performed based on the starting position and the updated set of historical sampling positions. When the position sampling stop condition is met, a parking path is generated based on the updated set of historical sampling positions.

[0200] A control module 15 is configured to control the vehicle to travel along the parking path based on the parking path.

[0201] In the embodiment, the sampling module comprises a region determining unit and a position sampling unit; the region determining unit is configured to determine a current sampling region and a plurality of extended sampling regions adjacent to the current sampling region based on a starting position of the vehicle and each historical sampling position in a set of historical sampling positions;

[0202] The position sampling unit is configured to perform position sampling on the current sampling region and each extended sampling region respectively based on a sampling probability corresponding to each sampling region, and obtain the plurality of current sampling positions.

[0203] In the embodiment, the sampling module further comprises a probability determining unit configured to determine a first sampling probability corresponding to the current sampling region, and determine a second sampling probability corresponding to each extended sampling region based on an area of each extended sampling region and the first sampling probability.

[0204] In the embodiment, the first determining module is configured to determine a target state corresponding to the target position based on the target position, calculate an initial cost value corresponding to each candidate arrival time of each current sampling position based on the target state and the candidate arrival time, and determine a minimum initial cost value as a target cost value of the current sampling position based on each initial cost value corresponding to the same current sampling position, so as to obtain a target cost value corresponding to each current sampling position.

[0205] In the embodiment, the second determining module is configured to determine a current sampling position corresponding to a minimum target cost value as the target current sampling position based on the target cost value of each current sampling position.

[0206] In the embodiment, the generating module is configured to determine whether a predetermined position sampling stop condition is met based on a distance between the target sampling position and the target position and / or a sampling wheel number.

[0207] In the embodiment, the control module is configured to determine a next driving position from the parking path based on a current driving position in real time, calculate a lateral error and a longitudinal error of the vehicle based on the current driving position and the next driving position, calculate a yaw rate of the vehicle based on the lateral error and the longitudinal error, calculate a front wheel steering angle of the vehicle based on the yaw rate, and control the vehicle to drive along the parking path based on the front wheel steering angle.

[0208] The autonomous parking control device in the embodiment can make the determination of the current sampling position more reasonable, and can subsequently filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate. The above process is repeated until the position sampling is completed, and the parking path can be generated based on the target sampling position obtained in each round of sampling (i.e., each historical sampling position in the historical sampling set). That is, the plurality of current sampling positions of the next path position point are determined by using the iterative sampling method, and the target sampling position of the next path position point is determined from the current sampling positions by combining the cost calculation, so that the determination of the target sampling position is more reasonable and accurate, and the parking path generated based on the target sampling position is more reasonable and accurate. Subsequently, autonomous parking can be accurately performed based on the parking path.

[0209] Another embodiment of the present application provides a storage medium storing a computer program, wherein the computer program is executed by a processor to implement the following method steps:

[0210] Step one, position sampling is performed based on a starting position of a vehicle and a historical sampling position set to obtain a plurality of current sampling positions;

[0211] Step two, a target cost value corresponding to each current sampling position is determined based on each current sampling position and a target position, respectively;

[0212] Step three, a target sampling position is determined based on the target cost values of the current sampling positions, and the target sampling position is added to the historical sampling position set to obtain an updated historical sampling position set;

[0213] Step four, it is determined whether a predetermined position sampling stop condition is met. When the position sampling stop condition is not met, the next round of position sampling is performed based on the starting position and the updated historical sampling position set, and when the position sampling stop condition is met, a parking path is generated based on the updated historical sampling position set;

[0214] Step five, the vehicle is controlled to drive according to the parking path based on the parking path.

[0215] The specific implementation process of the above method steps can be referred to the embodiments of any autonomous parking method described above, and will not be repeated here.

[0216] The storage medium in the present application determines a plurality of current sampling positions by randomly sampling according to the starting position and the set of historical sampling positions, so that the determination of the current sampling position is more reasonable, and a target sampling position can be selected from the plurality of current sampling positions by using the cost calculation method, and the target sampling position is used as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate. In this way, the cycle is repeated until the position sampling is completed, and the parking path can be generated based on the target sampling position obtained in each round of sampling (i.e., each historical sampling position in the set of historical sampling positions). That is, by using the iterative sampling method to determine a plurality of current sampling positions of the next path position point, and combining the cost calculation, the target sampling position as the next path position point is determined from the current sampling positions, so that the determination of the target sampling position is more reasonable and accurate, and the parking path generated based on the target sampling position is more reasonable and accurate, and subsequent autonomous parking can be accurately performed based on the parking path.

[0217] Another embodiment of the present application provides an electronic device, as shown in FIG. 5, which at least includes a memory 1 and a processor 2. The memory 1 stores a computer program, and the processor 2 implements the following method steps when executing the computer program on the memory 1.

[0218] Step one, position sampling is performed based on the starting position of the vehicle and the set of historical sampling positions to obtain a plurality of current sampling positions.

[0219] Step two, respectively based on each of the current sampling positions and the target position, a target cost value corresponding to each of the current sampling positions is determined.

[0220] Step three, based on the target cost values of the current sampling positions, a target sampling position is determined, and the target sampling position is added to the set of historical sampling positions to obtain an updated set of historical sampling positions.

[0221] Step four, it is determined whether a predetermined position sampling stop condition is met. When the position sampling stop condition is not met, the next round of position sampling is performed based on the starting position and the updated set of historical sampling positions, and when the position sampling stop condition is met, a parking path is generated based on the updated set of historical sampling positions.

[0222] Step five, the vehicle is controlled to travel according to the parking path based on the parking path.

[0223] The specific implementation process of the above method steps can be referred to the embodiments of any autonomous parking method described above, and the present embodiment will not be repeated here.

[0224] The electronic device in the application can determine the current sampling positions more reasonably by randomly sampling according to the starting position and the historical sampling position set in real time, and can subsequently filter out a target sampling position from the plurality of current sampling positions by using the cost calculation method, and use the target sampling position as a position point in the parking path, so that the determination of the target sampling position is more reasonable and accurate. This cycle is repeated until the position sampling is completed, and the parking path can be generated based on the target sampling positions obtained in each round of sampling (i.e., the historical sampling positions in the historical sampling set). That is, by using the iterative sampling method to determine the plurality of current sampling positions of the next path position point, and combining the cost calculation to determine the target sampling position as the next path position point from the current sampling positions, the determination of the target sampling position can be more reasonable and accurate, and the parking path generated based on the target sampling position can be more reasonable and accurate, and subsequent autonomous parking can be accurately performed based on the parking path.

[0225] The above embodiments are only exemplary embodiments of the application and are not intended to limit the application. The protection scope of the application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements to the application within the spirit and protection scope of the application, and such modifications or equivalent replacements shall also be considered to fall within the protection scope of the application.

Claims

1. An autonomous parking method, wherein, The method comprises the following steps: sampling positions based on a starting position of a vehicle and a set of historical sampling positions to obtain a plurality of current sampling positions; determining target cost values corresponding to the current sampling positions based on the current sampling positions and a target position respectively; determining a target sampling position based on the target cost values of the current sampling positions and adding the target sampling position to the set of historical sampling positions to obtain an updated set of historical sampling positions; determining whether a predetermined position sampling stop condition is met; when the position sampling stop condition is not met, performing the next round of position sampling based on the starting position and the updated set of historical sampling positions, and when the position sampling stop condition is met, generating a parking path based on the updated set of historical sampling positions; controlling the vehicle to travel along the parking path based on the parking path.

2. The method of claim 1, wherein, The method of sampling positions based on a starting position of a vehicle and a set of historical sampling positions to obtain a plurality of current sampling positions comprises the following steps: determining a current sampling region and a plurality of extended sampling regions adjacent to the current sampling region based on the starting position of the vehicle and each historical sampling position in the set of historical sampling positions; sampling positions in the corresponding current sampling region and each extended sampling region based on the sampling probability corresponding to each sampling region to obtain the plurality of current sampling positions.

3. The method of claim 2, wherein, Before the random sampling, the method further comprises determining the sampling probability corresponding to the current sampling region and each extended sampling region, which comprises the following steps: determining a first sampling probability corresponding to the current sampling region; determining a second sampling probability corresponding to each extended sampling region based on the area of each extended sampling region and the first sampling probability.

4. The method of claim 1, wherein, The method of determining target cost values corresponding to the current sampling positions based on the current sampling positions and a target position respectively comprises the following steps: determining a target state corresponding to the target position based on the target position; calculating initial cost values corresponding to a plurality of candidate arrival times based on the target state and the candidate arrival times corresponding to each current sampling position; determining a minimum initial cost value as the target cost value of the current sampling position based on the initial cost values corresponding to the same current sampling position to obtain target cost values corresponding to each current sampling position.

5. The method of claim 1, wherein, The method of determining a target sampling position based on the target cost values of the current sampling positions comprises the following steps: determining the current sampling position corresponding to the minimum target cost value as the target current sampling position based on the target cost values of the current sampling positions.

6. The method of claim 1, wherein, The method of determining whether a predetermined position sampling stop condition is met comprises the following steps: determining whether a predetermined position sampling stop condition is met based on the distance between the target sampling position and the target position and / or the number of sampling rounds.

7. The method of claim 1, wherein, The method of controlling the vehicle to travel along the parking path based on the parking path comprises the following steps: determining a next travel position from the parking path in real time based on the current travel position; calculating a lateral error and a longitudinal error of the vehicle based on the current driving position and the next driving position; calculating a yaw rate of the vehicle based on the lateral error and the longitudinal error; calculating a front wheel steering angle of the vehicle based on the yaw rate; controlling the vehicle based on the front wheel steering angle to drive the vehicle along the parking path.

8. A parking control device, wherein, comprising: a sampling module configured to sample positions based on a starting position of the vehicle and a set of historical sampling positions to obtain a plurality of current sampling positions; a first determining module configured to determine a target cost value corresponding to each of the current sampling positions based on each of the current sampling positions and a target position respectively; a second determining module configured to determine a target sampling position based on the target cost value of each of the current sampling positions, and add the target sampling position to the set of historical sampling positions to obtain an updated set of historical sampling positions; a generating module configured to determine whether a predetermined position sampling stop condition is met; when the position sampling stop condition is not met, performing a next round of position sampling based on the starting position and the updated set of historical sampling positions, and when the position sampling stop condition is met, generating a parking path based on the updated set of historical sampling positions; a control module configured to control the vehicle to drive along the parking path based on the parking path.

9. A storage medium, wherein, The storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the autonomous parking method in any one of claims 1-7.

10. An electronic device, comprising: The device at least comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the autonomous parking method in any one of claims 1-7 when executing the computer program on the memory.

Images