An automatic parking method, apparatus, electronic device and storage medium

Abstract

This application relates to the field of autonomous driving technology, and more particularly to an automatic parking method, apparatus, electronic device, and storage medium. Applied to four-wheel steering vehicles, the method includes: in response to a vehicle parking operation, acquiring a set of parking control parameters; controlling the vehicle to perform automatic parking based on the parking control parameters in the set; during the automatic parking process, acquiring the vehicle's current front wheel angle, current driving trajectory information, and planned driving trajectory information; determining the front-rear wheel ratio coefficient corresponding to the current front wheel angle, the front-rear wheel ratio coefficient indicating the ratio between the vehicle's rear wheel angle and front wheel angle; and controlling the vehicle's driving trajectory during the automatic parking process based on the angle ratio coefficient, driving trajectory information, and planned driving trajectory information. This application improves the efficiency of parking control parameter calibration and enhances the safety, stability, practicality, and comfort of automatic parking.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to an automatic parking method, device, electronic device, and storage medium. Background Technology

[0002] As cars become more and more common, the requirements for car safety and stability are getting higher and higher. Four-wheel steering cars have emerged to meet this need. Four-wheel steering cars can ensure driving stability at high speeds and reduce the turning radius at low speeds.

[0003] Currently, four-wheel steering vehicle control does not consider the influence of rear wheel steering angle, resulting in poor safety, stability, practicality, and comfort in automatic parking. Therefore, there is a need to provide an improved automatic parking method to enhance the safety, stability, practicality, and comfort of automatic parking. Summary of the Invention

[0004] In view of the above-mentioned problems in the prior art, this application provides an automatic parking method, device, electronic device and storage medium to improve the safety, stability, practicality and comfort of automatic parking.

[0005] To achieve the objective, the technical solution adopted in this application is:

[0006] On the one hand, this application provides an automatic parking method applied to a four-wheel steering vehicle, the method comprising:

[0007] In response to a vehicle parking operation, a parking control parameter set is acquired. The parking control parameter set is obtained by optimizing and calibrating a reference parking control parameter set based on a preset optimization algorithm. The reference parking control parameter set is obtained by extracting calibration parameters based on the parameter range of multiple parking control parameters.

[0008] The vehicle is automatically parked based on the parking control parameters in the parking control parameter set.

[0009] During the automatic parking process of the vehicle, the current front wheel turning angle, current driving trajectory information and planned driving trajectory information of the vehicle are acquired. The planned driving trajectory information represents the planned driving trajectory information of the vehicle on the target parking segment formed between the target parking start point and the target parking end point. The current driving trajectory information represents the actual trajectory information formed by the vehicle during the parking process.

[0010] Determine the front-to-rear wheel ratio coefficient corresponding to the current front wheel steering angle, wherein the front-to-rear wheel ratio coefficient indicates the ratio between the rear wheel steering angle and the front wheel steering angle of the vehicle;

[0011] The vehicle's driving trajectory during automatic parking is controlled based on the turning angle ratio coefficient, the driving trajectory information, and the planned driving trajectory information.

[0012] Furthermore, the acquisition of the parking control parameter set includes:

[0013] Obtain multiple simulated driving trajectories;

[0014] The reference parking control parameter set is used as the initial population of the preset optimization algorithm, and the control parameters in the parking control parameter set are set to correspond one-to-one with the genes in the individuals of the initial population.

[0015] Based on a preset optimization algorithm, each individual in the initial population is optimized to approach the optimal solution, thereby obtaining the target offspring population corresponding to each of the multiple simulated planned driving trajectories.

[0016] For each target offspring population corresponding to the plurality of simulated planned driving trajectories, a fitness evaluation process is performed on each of the simulated planned driving trajectories to obtain a trajectory fitness evaluation result for each of the plurality of target offspring populations. The trajectory fitness evaluation result indicates the degree of matching between the simulated planned driving trajectory corresponding to the target offspring population and the plurality of simulated planned driving trajectories.

[0017] Based on the trajectory fitness evaluation results, the optimal offspring population is selected from the multiple target offspring populations;

[0018] The optimal offspring population is determined as the parking control parameter set.

[0019] Furthermore, the optimization process based on a preset optimization algorithm, which approaches the optimal solution for each individual in the initial population, yields the target offspring populations corresponding to each of the multiple simulated planned driving trajectories, including:

[0020] Under the multiple simulated planned driving trajectories, the fitness evaluation process is performed on each individual in the initial population to obtain the evaluation value corresponding to each individual; the evaluation value indicates the degree of closeness between the individual and the optimal solution corresponding to the individual;

[0021] Based on the evaluation value, the initial population is subjected to probability extraction to obtain the first generation population;

[0022] The first generation population was subjected to crossover mutation to obtain the first offspring population.

[0023] The first offspring population was subjected to simulated annealing based on a preset cooling rate to obtain the second offspring population.

[0024] Using the second offspring population as the initial population, the fitness evaluation process, the probability extraction process, the crossover mutation process, and the simulated annealing process are executed sequentially in a loop until the preset number of loops is reached.

[0025] The population obtained when the loop execution reaches the preset number of loops is determined as the target offspring population corresponding to each of the multiple simulated planned driving trajectories.

[0026] Furthermore, the process of performing crossover mutation on the first generation population to obtain the first offspring population includes:

[0027] Perform crossover probability selection on the first generation population to determine the crossover individuals that need to be crossovered in the first generation population.

[0028] The crossover individuals are then crossovered to obtain the second generation population;

[0029] The second-generation population is subjected to mutation probability selection to determine the mutant individuals that need to be mutated in the second-generation population.

[0030] The mutated individuals are subjected to mutation treatment to obtain the first generation population.

[0031] Furthermore, the step of performing simulated annealing on the first offspring population based on a preset cooling rate to obtain the second offspring population includes:

[0032] The fitness of the first offspring population is evaluated to obtain the initial evaluation value of the first offspring population.

[0033] Obtain the initial temperature and the preset cooling rate; the initial temperature and the preset cooling rate are related to the individual's retention tolerance.

[0034] Each gene in the first offspring individuals of the first offspring population is updated with a preset step size to generate an updated first offspring individual corresponding to the first offspring population; the preset step size indicates whether the parameter corresponding to each gene in the first offspring individual is increased or decreased within a preset value range.

[0035] The fitness evaluation process is performed on the updated first offspring individual to obtain the update evaluation value corresponding to the updated first offspring individual;

[0036] Determine the difference between the updated evaluation value and the initial evaluation value;

[0037] The initial temperature, the preset cooling rate, and the difference in the evaluation value are evaluated and calculated to obtain the individual evaluation probability of any one of the first offspring individuals and the updated first offspring individuals.

[0038] The steps of updating the preset step size, evaluating fitness, subtracting, and calculating the evaluation are executed repeatedly until the preset number of iterations is reached. The second offspring population is then obtained based on the individual evaluation probability.

[0039] Further, the step of performing fitness evaluation processing on each target offspring population corresponding to each of the multiple simulated planned driving trajectories to obtain the trajectory fitness evaluation results for each of the multiple target offspring populations includes:

[0040] For each target offspring population corresponding to the multiple simulated planned driving trajectories, a fitness evaluation process is performed on each of the multiple simulated planned driving trajectories to obtain the evaluation value under the multiple simulated planned driving trajectories corresponding to each target offspring population.

[0041] The average evaluation value corresponding to each target offspring population is obtained by averaging the multiple evaluation values.

[0042] The multiple average evaluation values ​​are compared and processed to obtain the trajectory fitness evaluation results corresponding to each of the multiple target offspring populations;

[0043] The step of selecting the optimal offspring population from the plurality of target offspring populations based on the trajectory fitness evaluation results includes:

[0044] The target offspring population with the highest average evaluation value is selected from the trajectory fitness evaluation results;

[0045] The target offspring population with the highest average evaluation value is determined as the optimal offspring population.

[0046] Furthermore, controlling the vehicle's driving trajectory during automatic parking based on the turning angle ratio coefficient, the driving trajectory information, and the planned driving trajectory information includes:

[0047] The driving trajectory information and the planned driving trajectory information are compared and processed to obtain the driving trajectory deviation information;

[0048] The vehicle acceleration information corresponding to the driving trajectory deviation information is determined based on the longitudinal control model of the target vehicle.

[0049] Based on the lateral control model of the target vehicle, determine the driving trajectory deviation information and the steering wheel angle information of the vehicle corresponding to the steering angle ratio coefficient;

[0050] The vehicle's trajectory during automatic parking is controlled based on the acceleration information and the steering wheel angle information.

[0051] On the other hand, this application also provides an automatic parking control device, the device comprising:

[0052] Parking control parameter set acquisition module: used to acquire a parking control parameter set in response to vehicle parking operation. The parking control parameter set is obtained by optimizing and calibrating a reference parking control parameter set based on a preset optimization algorithm. The reference parking control parameter set is obtained by extracting calibration parameters based on the parameter range of multiple parking control parameters.

[0053] Automatic parking control module: used to control the vehicle to perform automatic parking based on the parking control parameters in the parking control parameter set;

[0054] Information acquisition module: used to acquire the vehicle's current front wheel turning angle, current driving trajectory information and planned driving trajectory information during the automatic parking process of the vehicle. The planned driving trajectory information represents the planned driving trajectory information of the vehicle on the target parking segment formed between the target parking start point and the target parking end point. The current driving trajectory information represents the actual trajectory information formed by the vehicle during the parking process.

[0055] Front and rear wheel ratio coefficient acquisition module: used to determine the front and rear wheel ratio coefficient corresponding to the current front wheel steering angle, wherein the front and rear wheel ratio coefficient indicates the ratio between the rear wheel steering angle and the front wheel steering angle of the vehicle;

[0056] The automatic parking control module is also used to control the vehicle's driving trajectory during the automatic parking process based on the turning angle ratio coefficient, the current driving trajectory information, and the planned driving trajectory information.

[0057] On the other hand, this application also provides an electronic device, the device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the automatic parking method as described above.

[0058] On the other hand, this application also provides a computer storage medium storing at least one instruction or at least one program, which is loaded and executed by a processor to implement the automatic parking method described above.

[0059] The beneficial effects of the technical solution in this application are:

[0060] This application obtains a parking control parameter set in response to a vehicle parking operation; controls the vehicle to perform automatic parking based on the parking control parameters in the parking control parameter set; during the automatic parking process, it obtains the vehicle's current front wheel angle, current driving trajectory information, and planned driving trajectory information. The planned driving trajectory information represents the planned driving trajectory of the vehicle on the target parking segment formed between the target parking start point and the target parking end point, and the current driving trajectory information represents the actual trajectory formed by the vehicle during the parking process; it determines the front-rear wheel ratio coefficient corresponding to the current front wheel angle; and controls the vehicle's driving trajectory during the automatic parking process based on the angle ratio coefficient, driving trajectory information, and planned driving trajectory information. Therefore, this application obtains optimized parking control parameters based on a preset optimization algorithm. Under diverse parking environments, the vehicle can be controlled by a single optimized parking control parameter, enabling rapid calibration of the parking control parameter and improving the efficiency of the calibration. At the same time, this application considers the ratio between the front wheel angle and the rear wheel angle, and controls the vehicle's driving trajectory during automatic parking based on the angle ratio coefficient, driving trajectory information, and planned driving trajectory information, thereby improving the safety, stability, practicality, and comfort of automatic parking. Attached Figure Description

[0061] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0063] Figure 2 This is a schematic diagram of the control principle of a vertical PID controller provided in an embodiment of this application;

[0064] Figure 3 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0065] Figure 4 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0066] Figure 5 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0067] Figure 6 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0068] Figure 7 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0069] Figure 8 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0070] Figure 9 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0071] Figure 10a This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0072] Figure 10b This is a flowchart illustrating an automatic parking method provided in an embodiment of this application;

[0073] Figure 11 This is a schematic block diagram of the structure of an automatic parking control device provided in an embodiment of this application;

[0074] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0075] This application discloses an automatic parking method, apparatus, electronic device and storage medium, control method and control device. The automatic parking method based on neural network and genetic algorithm has a simple structure, reduces power consumption, and improves heating efficiency.

[0076] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0077] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0078] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0079] Simulated Annealing Genetic Algorithm: This is a heuristic random search algorithm based on the Monte Carlo iterative solution method. It simulates the thermal equilibrium problem of solid material annealing process and the similarity between the problem and the random search optimization problem to find the global optimum or near-global optimum.

[0080] Genetic Algorithm: In computer science and operations research, a genetic algorithm is a metaheuristic algorithm inspired by the process of natural selection, belonging to the broad category of evolutionary algorithms. Genetic algorithms typically rely on biologically inspired operators, such as mutation, crossover, and selection, to generate high-quality solutions to optimization and search problems.

[0081] The following combination Figure 1 This application discloses an automatic parking method applicable to four-wheel steering vehicles. Please refer to [link / reference needed]. Figure 1 , Figure 1 This is a flowchart illustrating an automatic parking method provided in an embodiment of this application. This application provides method operation steps as shown in the embodiment or flowchart, but based on conventional or non-inventive labor, more or fewer operation steps may be included. The order of steps listed in the embodiment is merely one possible execution order among many, and does not represent the only execution order. In actual device, system, or equipment products, the method can be executed sequentially according to the embodiment or the accompanying drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment). Specifically, as shown... Figure 1 As shown, the method may include:

[0082] S101: In response to a vehicle parking operation, obtain the parking control parameter set.

[0083] The parking control parameter set is obtained by optimizing and calibrating the reference parking control parameter set based on a preset optimization algorithm. The reference parking control parameter set is obtained by extracting calibration parameters based on the parameter range of multiple parking control parameters.

[0084] In some embodiments, the preset optimization algorithm may be a simulated annealing genetic algorithm, or a genetic algorithm and particle swarm optimization algorithm, which can optimize the reference parking control parameters.

[0085] In some embodiments, the parking control parameter set includes lateral control parameters and longitudinal control parameters. Specifically, the lateral control parameters can be control parameters of an LQR controller, and the longitudinal control parameters can be control parameters of a PID controller.

[0086] For example, when the lateral control parameters in the parking control parameter set are control parameters of an LQR controller and the longitudinal control parameters are control parameters of a PID controller, the parking control parameter set includes: position proportional coefficient (K... SP ), position integral coefficient (K)SI ), positional differential coefficients (K) SD ), speed proportionality coefficient (K) VP ), velocity integral coefficient (K) VI ), velocity differential coefficient (K) VD ), lateral error coefficient (q1), lateral error rate of change coefficient (q2), heading angle error coefficient (q3), and heading angle error rate of change coefficient (q4), where K SP K SI K SD K VP K VI K VD q1, q2, q3, and q4 are the control parameters for the PID controller, while q1, q2, q3, and q4 are the control parameters for the LQR controller.

[0087] Specifically, please refer to the control algorithm of the longitudinal PID controller. Figure 2 And the following formulas 1 and 2:

[0088]

[0089]

[0090] The longitudinal PID controller includes a position PID controller and a speed PID controller. Formula 1 is the control algorithm for the position PID controller, and Formula 2 is the control algorithm for the speed PID control parameters. In this formula, u1 is the speed compensation amount, e1 is the position error (the deviation between the actual position and the planned position during parking), u2 is the longitudinal acceleration of the vehicle, e2 is the speed error after speed compensation, e3 is the speed error (the deviation between the actual speed and the planned speed during parking), k is the discrete time point, k represents the parking end time of the kth parking control cycle, k-1 represents the parking end time of the (k-1)th parking control cycle, which is the parking start time of the kth parking control cycle.

[0091] In practical applications, during the parking control cycle, the position error e1 and the position proportional coefficient K are... SP Position integral coefficient K SI and positional differential coefficients K SD As input to the position PID controller, the speed compensation amount u1 is calculated based on Formula 1, where u1 is the speed compensation amount for the k-th parking control cycle; the speed compensation amount u1, speed error e3, speed error e2 after speed compensation, and speed proportional coefficient K are then used as inputs to the position PID controller. VP Speed ​​integral coefficient K VI velocity differential coefficient K VDAs input to the speed PID controller, the vehicle's longitudinal acceleration u2 is calculated based on Formula 2. u2 represents the vehicle's longitudinal acceleration in the k-th parking control cycle, thereby controlling the output of the vehicle's longitudinal acceleration. Specifically, the vehicle's longitudinal acceleration can be the longitudinal acceleration of the vehicle's throttle.

[0092] Specifically, for the control algorithm of the horizontal LQR controller, please refer to Formulas 3, 4, 5 and 6 below.

[0093] In this embodiment of the application, based on the vehicle dynamics model, the steering angle ratio coefficient between the rear wheel steering angle and the front wheel steering angle is referenced to improve the vehicle dynamics model, resulting in the following model (Formula 3):

[0094]

[0095] Where y is the lateral movement distance. C is the heading angle (the angle between the direction the car is heading and the tangent to the circle at that point). αf C represents the lateral stiffness of the vehicle's front wheels. αr Let m be the lateral stiffness of the rear wheel of the vehicle, and I be the total vehicle mass. z Let l be the moment of inertia of the vehicle about the z-axis. f For the front overhang length, l r V is the rear overhang length. x δ is the longitudinal speed of the vehicle, k is the ratio coefficient of the rear wheel angle to the front wheel angle, k is related to the speed and the front wheel angle, and δ is the steering wheel angle.

[0096] Based on Equation 3, the dynamic model of steering wheel control is improved to obtain the following model (Equation 4):

[0097]

[0098] Where e1 is the lateral error. The rate of change of lateral error, where e2 is the heading angle error. Rate of change of heading angle C is the desired rate of change of heading angle. αf C represents the lateral stiffness of the vehicle's front wheels. αr Let m be the lateral stiffness of the rear wheel of the vehicle, and I be the total vehicle mass. z Let l be the moment of inertia of the vehicle about the z-axis. f For the front overhang length, l r V is the rear overhang length. x δ is the longitudinal speed of the vehicle, k is the ratio coefficient of the rear wheel angle to the front wheel angle, k is related to the speed and the front wheel angle, and δ is the steering wheel angle.

[0099] Equation 4 is simplified to Equation 5, as shown below:

[0100]

[0101] in,

[0102] By using a lateral LQR controller, the output J of the energy function is minimized by adjusting Q, thereby determining the δ corresponding to the minimum output J, and thus controlling the steering wheel angle of the vehicle. The energy function (Equation 6) is shown below:

[0103]

[0104] Where Q is a positive semi-definite matrix. R is a positive definite matrix, R = [1].

[0105] S102: The vehicle is automatically parked based on the parking control parameters in the parking control parameter set. It should be noted that the parking control parameters in step S102 are similar to those in step S101. For example, when the lateral control parameters in the parking control parameter set are the control parameters of the LQR controller and the longitudinal control parameters are the control parameters of the PID controller, it can be based on K... SP K SI K SD K VP K VI K VD q1, q2, q3, and q4 control the vehicle's longitudinal acceleration and steering wheel angle to enable the vehicle to perform automatic parking.

[0106] S103: During the automatic parking process, obtain the vehicle's current front wheel angle, current driving trajectory information, and planned driving trajectory information.

[0107] The planned driving trajectory information represents the planned driving trajectory of the vehicle on the target parking segment formed between the target parking start point and the target parking end point. The current driving trajectory information represents the actual trajectory information formed by the vehicle during the parking process. The current driving trajectory information includes, but is not limited to, the kinematic information of each trajectory point on the current driving trajectory, as well as the velocity, acceleration, heading angle, and rate of change of heading angle at the trajectory points. The planned driving trajectory information includes, but is not limited to, the kinematic information of each planned trajectory point on the planned driving trajectory, as well as the planned velocity, planned acceleration, planned heading angle, and rate of change of heading angle at the planned trajectory points.

[0108] In some embodiments, the current driving trajectory information may include information such as the current vehicle speed and position information that can characterize the actual trajectory formed by the current vehicle during the parking process. The target parking segment may be the entire road segment from the target parking start point to the target parking end point, or it may be a segmented road segment corresponding to one parking control cycle between the target parking start point and the target parking end point.

[0109] S104: Determine the front-to-rear wheel ratio coefficient corresponding to the current front wheel steering angle.

[0110] The front-to-rear wheel ratio indicates the ratio between the steering angle of the rear wheels and the steering angle of the front wheels of a vehicle.

[0111] In some embodiments, the front-to-rear wheel ratio coefficient is obtained based on the correspondence between the front wheel steering angle and the front-to-rear wheel ratio coefficient. For example, the correspondence between the front wheel steering angle and the front-to-rear wheel ratio coefficient is shown in the table below:

[0112]

[0113] S105: Controls the vehicle's trajectory during automatic parking based on the turning angle ratio coefficient, driving trajectory information, and planned driving trajectory information.

[0114] This application obtains optimized parking control parameters based on a preset optimization algorithm. Under diverse parking environments, the vehicle can be controlled to park using a single optimized parking control parameter, enabling rapid calibration of the parking control parameter and improving the efficiency of calibration. Simultaneously, this application considers the ratio between the front and rear wheel steering angles, controlling the vehicle's trajectory during automatic parking based on the steering angle ratio coefficient, driving trajectory information, and planned driving trajectory information, thereby improving the safety, stability, practicality, and comfort of automatic parking.

[0115] In some embodiments, please refer to Figure 9 Step S105 may include S701-S704.

[0116] S701: Compare and process the current driving trajectory information and the planned driving trajectory information to obtain the driving trajectory deviation information.

[0117] The current driving trajectory information includes, but is not limited to, the kinematic information of each trajectory point on the current driving trajectory, such as velocity, acceleration, heading angle, and rate of change of heading angle at each trajectory point. The planned driving trajectory information includes, but is not limited to, the kinematic information of each planned trajectory point on the planned driving trajectory, such as planned velocity, planned acceleration, planned heading angle, and rate of change of heading angle at each planned trajectory point. It should be noted that the current driving trajectory information can be determined based on the vehicle's actual driving position at various times. The difference between the current driving trajectory information and the planned driving trajectory information can be calculated to obtain the deviation information between the two. The driving trajectory deviation information can include the position difference information between the current driving trajectory and the planned driving trajectory, the velocity difference information between the current driving trajectory and the planned driving trajectory, and the heading angle difference information between the current driving trajectory and the planned driving trajectory.

[0118] S702: Determine the vehicle acceleration information corresponding to the trajectory deviation information based on the longitudinal control model of the target vehicle.

[0119] In some embodiments, the longitudinal control model can be a longitudinal PID controller, please refer to Equations 1 and 2, and then based on the driving trajectory information, the planned driving trajectory information, and the K value in the parking control parameter set. SP K SI K SD K VP K VI and K VD The vehicle acceleration information corresponding to the deviation in the driving trajectory is obtained. It should be noted that the vehicle acceleration information refers to the longitudinal acceleration of the vehicle.

[0120] S703: Based on the lateral control model of the target vehicle, determine the vehicle's steering wheel angle information corresponding to the driving trajectory deviation and the steering angle ratio coefficient.

[0121] In some embodiments, the lateral control model can be a lateral LQR controller. For details, please refer to Formulas 3, 4, 5 and 6. Then, based on the driving trajectory information, the planned driving trajectory information, the steering angle proportional coefficient and q1, q2, q3 and q4 in the parking control parameter set, the driving trajectory deviation information and the steering wheel angle information corresponding to the steering angle proportional coefficient are obtained. The steering wheel angle information is the steering wheel angle adjustment angle of the vehicle.

[0122] S704: Controls the vehicle's trajectory during automatic parking based on acceleration and steering wheel angle information.

[0123] In some embodiments, please refer to Figure 3 Step S101 includes S201-S205:

[0124] S201: Obtain multiple simulated driving trajectories.

[0125] In some embodiments, multiple simulated planned driving trajectories are planned driving trajectories for simulated parking based on multiple different target parking starting points to different target parking ending points.

[0126] S202: Use the reference parking control parameter set as the initial population for the preset optimization algorithm.

[0127] In some embodiments, step S202 includes: obtaining the value range of each control parameter in the reference parking control parameter set; randomly assigning values ​​to the control parameters within the value range of each parameter to form a reference parking control parameter set; and using the reference parking control parameter set as the initial population of a preset optimization algorithm.

[0128] It should be noted that the control parameters in the parking control parameter set are set in a one-to-one correspondence with the genes in the individuals of the initial population. Each individual corresponds to a first preset number of genes. When the lateral control parameters in the parking control parameter set are those of the LQR controller and the longitudinal control parameters are those of the PID controller, the number of genes is the same as the number of control parameters in the parking control parameter set. For example, the first preset number can be 10. The value range of each control parameter (gene) in the control parameter set can be determined empirically, or the maximum value range of each control parameter (gene) can be determined according to the vehicle type.

[0129] For example, when the lateral control parameters are those of an LQR controller and the longitudinal control parameters are those of a PID controller, the number of control parameters in the reference parking control parameter set can be 10, and the number of control parameters in the reference parking control parameter set can be K. Sp K SI K SD K vp K vI K VD The value ranges of the control parameters q1, q2, q3, and q4 in the reference parking control parameter set are as follows: K Sp The value of K can range from 0 to 10 (0 is not allowed). vP The value of K can range from 0 to 10 (0 is not allowed). VD The value of K can range from 0 to 0.1. SD The value of K can range from 0 to 0.1. SI The value of K can range from 0 to 0.1. VIThe value range of q1 can be 0 to 0.1, the value range of q2 can be 0 to 1, the value range of q3 can be 0 to 10 (0 is not allowed), and the value range of q4 can be 0 to 1.

[0130] In some embodiments, the initial population includes a second preset number of individuals, that is, the initial population includes a second preset number of reference parking control parameter sets, which can be 20-50.

[0131] S203: Based on the preset optimization algorithm, each individual in the initial population is optimized to approach the optimal solution, resulting in the target offspring population corresponding to each of the multiple simulated planned driving trajectories.

[0132] Specifically, the preset optimization algorithm is constructed based on the simulated annealing genetic algorithm, and then performs optimization processing on each individual in the initial population to approach the optimal solution.

[0133] In some embodiments, please refer to Figure 4 Step S203 includes steps S301-S306:

[0134] S301: Under multiple simulated planned driving trajectories, the fitness of each individual in the initial population is evaluated to obtain the evaluation value corresponding to each individual.

[0135] The evaluation value indicates the degree of closeness between an individual and its corresponding optimal solution. It should be noted that the fitness evaluation process can be calculated using the fitness function formula 7 above, and will not be elaborated further here.

[0136] S302: Based on the evaluation value, the initial population is subjected to probability extraction to obtain the first generation population.

[0137] This application obtains an optimized first-generation population by performing probability extraction based on evaluation values ​​on the initial population.

[0138] In some embodiments, the probability extraction process can be carried out by using a roulette wheel method to select and generate the first generation population according to probability. For details, please refer to the following population selection probability calculation formula (Formula 8):

[0139]

[0140] Where n represents the nth individual in the initial population, F(c) n P represents the fitness function value obtained by the nth individual through fitness evaluation. n This represents the probability that the nth element is selected as part of the first generation population.

[0141] Specifically, fitness evaluation is performed on each individual in the initial population to obtain an evaluation value for each individual. Based on the evaluation value, the probability of each individual being selected for the first generation population is calculated using Formula 8. Then, based on the probability of each individual being selected for the first generation population, a roulette wheel selection method is used to randomly select the first generation population from the initial population. This is understandable, P... n The larger the size, the greater the probability of being selected as part of the first generation population.

[0142] S303: The first generation population is obtained by performing crossover and mutation treatment on the first generation population. This application can optimize the first generation population to obtain the first generation population through various crossover and mutation treatments.

[0143] In some embodiments, crossover and mutation processing includes crossover processing and mutation processing.

[0144] In some embodiments, crossover processing includes: single-point crossover, two-point crossover, uniform crossover, and arithmetic crossover. Single-point crossover involves randomly setting only one crossover point in the individual's encoding string, and then exchanging portions of the chromosomes of the two paired individuals at that point. Two-point crossover involves randomly setting two crossover points in the individual's encoding string, and then exchanging partial genes. Uniform crossover involves exchanging genes at each locus of the two paired individuals with the same crossover probability, thus forming two new individuals. Arithmetic crossover generates two new individuals through a linear combination of two individuals. The objects of this operation are generally individuals represented by floating-point codes.

[0145] In some embodiments, mutation processing includes: basic bit mutation, uniform mutation, boundary mutation, non-uniform mutation, and Gaussian approximation mutation. Basic bit mutation refers to performing mutation operations on a single bit or several randomly selected bits in the individual coding string with a mutation probability, based solely on the value at the locus. Uniform mutation refers to replacing the original gene values ​​at each locus in the individual coding string with random numbers uniformly distributed within a certain range, with a small probability. Boundary mutation refers to randomly selecting one of the two corresponding boundary gene values ​​at a locus to replace the original gene value. Non-uniform mutation refers to randomly perturbing the original gene values, using the perturbation result as the new gene value after mutation. Performing mutation operations on each locus with the same probability is equivalent to a slight change in the entire solution vector within the solution space. Gaussian approximation mutation refers to using the average value of the sign mean P and the variance σ during the mutation operation. 2 A random number from a normal distribution is used to replace the original gene value.

[0146] In some embodiments, please refer to Figure 5 Step S303 includes S401-S404:

[0147] S401: Perform crossover probability selection on the first generation population to determine the crossover individuals that need to be crossovered in the first generation population.

[0148] In some embodiments, the crossover probability P c It can be less than or equal to 0.1, for example, in P c =0.1. If the first generation population has 20 individuals, the number of individuals that need to be crossed over is 2.

[0149] S402: Cross over the individuals to obtain the second generation population.

[0150] In one specific embodiment, a single-point crossover process is performed on the crossover individuals to obtain the second-generation population.

[0151] S403: Perform mutation probability selection on the second-generation population to determine the mutant individuals that need to be mutated in the second-generation population.

[0152] In some embodiments, the mutation probability P m It can be less than or equal to 0.05, for example, in P c =0.05, with 20 individuals in the second generation population, the number of individuals that need to be mutated is 1.

[0153] S404: Perform mutation processing on the mutated individuals to obtain the first generation population.

[0154] In one specific embodiment, the mutated individuals are subjected to uniform mutation to obtain the first generation population.

[0155] This application obtains the first generation population by performing crossover and mutation processing on the first generation population, thereby updating and changing the first generation population on a small scale, and thus providing a basis for obtaining an optimal set of parking control parameters.

[0156] S304: Simulated annealing is performed on the first offspring population based on a preset cooling rate to obtain the second offspring population. This application optimizes the first offspring population through simulated annealing to obtain the second offspring population.

[0157] In some embodiments, the preset cooling rate can be 0.95-0.98, and the preset cooling rate is related to the individual's retention tolerance.

[0158] In some embodiments, please refer to Figure 6 Step S304 includes S501-S507:

[0159] S501: Perform fitness evaluation on the first generation population to obtain the initial evaluation value corresponding to the first generation population.

[0160] It should be noted that the fitness evaluation process can be calculated using the fitness function formula 7 above, and will not be elaborated further here.

[0161] S502: Obtain the initial temperature and preset cooling rate; the initial temperature and preset cooling rate are related to the individual's retention tolerance.

[0162] In some embodiments, the initial temperature may be determined based on specific circumstances (e.g., population size, fitness function value, and vehicle type), without limitation herein.

[0163] S503: Perform a preset step size update process on each gene in the first generation individuals of the first generation population to generate the updated first generation individuals corresponding to the first generation population; the preset step size indicates that the parameters corresponding to each gene in the first generation individuals are increased or decreased within a preset value range.

[0164] S504: Perform fitness evaluation on the updated first-generation individuals to obtain the update evaluation value corresponding to the updated first-generation individuals. It should be noted that the fitness evaluation can be calculated using the fitness function formula 7 above, and will not be elaborated here.

[0165] S505: Determine the difference between the updated evaluation value and the initial evaluation value.

[0166] S506: Perform evaluation calculations on the initial temperature, preset cooling rate, and evaluation value difference to obtain the individual evaluation probability of any individual among the first offspring and the updated first offspring.

[0167] In a specific embodiment, the individual evaluation probability can be calculated based on the following evaluation calculation formulas (Formulas 9 and 10):

[0168]

[0169] T i =αT i-1 ,α∈(0,1) (Formula 10);

[0170] Among them, P ni Let ΔF(c) represent the individual evaluation probability value of the nth individual in the i-th simulated annealing; ni α represents the sum of the i-th evaluation value and the difference between the (i-1)-th evaluation value and the (i-1)-th evaluation value of the n-th individual; α is the preset cooling rate; T0 is the initial temperature; Ti represents the annealing temperature of the i-th simulated annealing process.

[0171] S507: Repeatedly execute the steps of update processing, fitness evaluation processing, difference processing and evaluation calculation processing with a preset step size until the preset number of iterations is reached, and obtain the second generation population based on the individual evaluation probability.

[0172] This application updates each gene in the first generation of individuals in the first generation population with a preset step size, and judges the changes in the fitness evaluation value of each individual before and after the update. It then expands and changes each gene in a small range, thereby optimizing each individual and providing a basis for obtaining an optimal set of parking control parameters.

[0173] S305: Using the second offspring population as the initial population, perform fitness evaluation, probability extraction, crossover mutation, and simulated annealing processes in a cyclical manner until the preset number of cycles is reached.

[0174] In some embodiments, the preset number of cycles can be 20-50.

[0175] S306: The population obtained after the loop execution reaches a preset number of iterations is determined as the target offspring population corresponding to each of the multiple simulated planned driving trajectories. This application repeatedly optimizes the population through multiple iterations to obtain an optimized target offspring population, optimized parking control parameters, and improves the efficiency of parking control parameter calibration.

[0176] S204: For each target offspring population corresponding to each of the multiple simulated planned driving trajectories, perform fitness evaluation processing for each simulated planned driving trajectory to obtain the trajectory fitness evaluation results for each of the multiple target offspring populations. The trajectory fitness evaluation results indicate the degree of matching between the simulated planned driving trajectory corresponding to the target offspring population and the multiple simulated planned driving trajectories.

[0177] Fitness evaluation can be obtained by calculating the fitness function, which can be represented by the following formula 7:

[0178]

[0179] Where F(c) is the fitness function value, representing the degree of optimization of an individual; the larger the F(c) value, the closer the corresponding individual is to the optimal solution; Q f Q d and R d These are non-negative vector coefficients, specifically set according to the vehicle's characteristics (e.g., vehicle type); z t z represents the actual driving trajectory information at time t. ref，t w represents the simulated planned driving trajectory information at time t. tw represents the control quantity output at time t. t+int This represents the control quantity of the output in the next control cycle at time t. For example, the control cycle can be 10 milliseconds.

[0180] Specifically,

[0181] Among them, v x For the speed of the vehicle, The vehicle's heading angle, where x represents the distance traveled in the direction of travel and y represents the distance traveled along the tangent of the travel trajectory.

[0182] Specifically, w t =[a,δ] T w t+int =[a,δ] T ;

[0183] Where a is the longitudinal acceleration of the vehicle, and δ is the steering wheel angle of the vehicle.

[0184] It should be noted that the above z t The target offspring population can be generated by performing simulated parking analysis on a pre-set simulation software using the target offspring population as parking control parameters, or it can be obtained by performing parking tests on a real vehicle using the target offspring population as parking control parameters; neither is limited here. Specifically, the pre-set simulation software can be CarSim.

[0185] S205: Based on the trajectory fitness evaluation results, the optimal offspring population is selected from multiple target offspring populations. It should be noted that the individuals in the optimal offspring population have the highest degree of similarity to the optimal solution.

[0186] In some embodiments, please refer to Figure 7 Step S204 includes S601-S603:

[0187] S601: For each target offspring population corresponding to each of the multiple simulated planned driving trajectories, perform fitness evaluation processing for each simulated planned driving trajectory to obtain the evaluation value for each target offspring population under the multiple simulated planned driving trajectories. It should be noted that the fitness evaluation processing can be calculated using the fitness function formula 7 above, and will not be elaborated here.

[0188] S602: Average the multiple evaluation values ​​corresponding to each target offspring population to obtain the average evaluation value corresponding to each target offspring population.

[0189] S603: Compare and process multiple average evaluation values ​​to obtain trajectory fitness evaluation results for each of the multiple target offspring populations. The trajectory fitness evaluation results characterize the degree of preference of the target offspring population.

[0190] Accordingly, please refer to Figure 8 Step S205 includes S604-S605:

[0191] S604: Select the target offspring population with the highest average evaluation value from the trajectory fitness evaluation results.

[0192] S605: The target offspring population with the highest average evaluation value is determined as the optimal offspring population.

[0193] This application evaluates the fitness of each target offspring population by performing multiple simulated planned driving trajectories, and obtains the average evaluation value of each target offspring population on multiple simulated planned driving trajectories. By comparing the average evaluation values, the target offspring population with the largest average evaluation value is determined as the optimal offspring population, taking into account the optimization of parking control parameters under diverse parking environments.

[0194] S206: Determine the optimal offspring population as the parking control parameter set.

[0195] This application obtains optimized parking control parameters through simulated annealing genetic algorithm. Under the premise of diverse parking environments, the vehicle can be controlled by a single optimized parking control parameter, which can quickly calibrate the parking control parameter and improve the efficiency of parking control parameter calibration.

[0196] The automatic parking method of this application is described below with reference to specific applications. Figure 10a The methods may include:

[0197] S1: Obtain the reference parking control parameter set.

[0198] S2: Use the reference parking control parameter set as the initial population for the preset optimization algorithm.

[0199] S3: Under multiple simulated planned driving trajectories, the fitness of each individual in the initial population is evaluated to obtain the evaluation value corresponding to each individual.

[0200] S4: Based on the evaluation value, perform probability extraction on the initial population to obtain the first generation population.

[0201] S5: Perform crossover probability selection on the first generation population to determine the crossover individuals that need to be crossovered in the first generation population.

[0202] S6: Perform crossover on the crossover individuals to obtain the second generation population.

[0203] S7: Perform mutation probability selection on the second-generation population to determine the mutant individuals that need to be mutated in the second-generation population.

[0204] S8: Perform mutation processing on the mutated individuals to obtain the first generation population.

[0205] S9: Perform fitness evaluation on the first generation population to obtain the initial evaluation value corresponding to the first generation population.

[0206] S10: Obtain the initial temperature and preset cooling rate.

[0207] S11: Perform a preset step size update process on each gene in the first generation individuals of the first generation population to generate the updated first generation individuals corresponding to the first generation population.

[0208] S12: Perform fitness evaluation on the updated first offspring individuals to obtain the update evaluation value corresponding to the updated first offspring individuals.

[0209] S13: Determine the difference between the updated evaluation value and the initial evaluation value.

[0210] S14: Perform evaluation calculations on the initial temperature, preset cooling rate, and evaluation value difference to obtain the individual evaluation probability of any individual among the first offspring and the updated first offspring.

[0211] S15: Repeatedly execute the steps of update processing, fitness evaluation processing, difference processing and evaluation calculation processing with a preset step size until the preset number of iterations is reached, and obtain the second generation population based on the individual evaluation probability.

[0212] Further, please refer to Figure 10b The methods also include:

[0213] S16: Using the second offspring population as the initial population, perform fitness evaluation, probability extraction, crossover mutation, and simulated annealing processes in a cyclical manner until the preset number of cycles is reached.

[0214] S17: The population obtained when the loop execution count reaches the preset number of loops is determined as the target offspring population corresponding to each of the multiple simulated planned driving trajectories.

[0215] S18: For each target offspring population in the target offspring population corresponding to each of the multiple simulated planned driving trajectories, perform fitness evaluation processing for each simulated planned driving trajectory in the multiple simulated planned driving trajectories to obtain the evaluation value under the multiple simulated planned driving trajectories corresponding to each target offspring population.

[0216] S19: Average the multiple evaluation values ​​corresponding to each target offspring population to obtain the average evaluation value corresponding to each target offspring population.

[0217] S20: Compare and process multiple average evaluation values ​​to obtain the trajectory fitness evaluation results for each of the multiple target offspring populations.

[0218] S21: Select the target offspring population with the highest average evaluation value from the trajectory fitness evaluation results.

[0219] S22: The target offspring population with the highest average evaluation value is determined as the optimal offspring population.

[0220] S23: Determine the optimal offspring population as the parking control parameter set.

[0221] S24: In response to a vehicle parking operation, automatically park the vehicle based on parking control parameters in the parking control parameter set.

[0222] S25: During the automatic parking process, obtain the vehicle's current front wheel angle, current driving trajectory information, and planned driving trajectory information.

[0223] S26: Determine the front-to-rear wheel ratio coefficient corresponding to the current front wheel steering angle.

[0224] S27: Compare and process the driving trajectory information and the planned driving trajectory information to obtain the driving trajectory deviation information.

[0225] S28: Determine the vehicle acceleration information corresponding to the trajectory deviation information based on the longitudinal control model of the target vehicle.

[0226] S29: Based on the lateral control model of the target vehicle, determine the vehicle's steering wheel angle information corresponding to the driving trajectory deviation and the steering angle ratio coefficient.

[0227] S30: Controls the vehicle's trajectory during automatic parking based on acceleration and steering wheel angle information.

[0228] In summary, the beneficial effects of the technical solution of this application are:

[0229] This application obtains optimized parking control parameters based on a preset optimization algorithm. Under diverse parking environments, the vehicle can be controlled to park using a single optimized parking control parameter, thus improving the efficiency of parking control parameter calibration. At the same time, this application considers the ratio between the front wheel angle and the rear wheel angle, and controls the vehicle's driving trajectory during automatic parking based on the angle ratio coefficient, driving trajectory information, and planned driving trajectory information, thereby improving the safety, stability, practicality, and comfort of automatic parking.

[0230] On the other hand, embodiments of this application also provide an automatic parking control device, please refer to... Figure 11 The device includes:

[0231] Parking control parameter set acquisition module 11: Used to acquire parking control parameter set in response to vehicle parking operation. The parking control parameter set is obtained by optimizing and calibrating the reference parking control parameter set based on a preset optimization algorithm. The reference parking control parameter set is obtained by calibrating parameters based on the parameter range of multiple parking control parameters.

[0232] Automatic parking control module 12: used to control the vehicle to perform automatic parking based on the parking control parameters in the parking control parameter set.

[0233] Information acquisition module 13: used to acquire the vehicle's current front wheel angle, current driving trajectory information and planned driving trajectory information during the automatic parking process. The planned driving trajectory information represents the planned driving trajectory information of the vehicle on the target parking segment formed between the target parking start point and the target parking end point. The current driving trajectory information represents the actual trajectory information formed by the vehicle during the parking process.

[0234] Front and rear wheel ratio coefficient acquisition module 14: used to determine the front and rear wheel ratio coefficient corresponding to the current front wheel steering angle. The front and rear wheel ratio coefficient indicates the ratio between the rear wheel steering angle and the front wheel steering angle of the vehicle.

[0235] The automatic parking control module 11 is also used to control the vehicle's driving trajectory during the automatic parking process based on the turning angle ratio coefficient, current driving trajectory information and planned driving trajectory information.

[0236] In some embodiments, the apparatus further includes:

[0237] Simulated driving trajectory acquisition module: used to acquire multiple simulated driving trajectories.

[0238] Initial Population Acquisition Module: This module uses the reference parking control parameter set as the initial population for the preset optimization algorithm. The control parameters in the parking control parameter set are set to correspond one-to-one with the individuals in the initial population.

[0239] The target offspring population acquisition module is used to optimize each individual in the initial population based on a preset optimization algorithm to approach the optimal solution, thereby obtaining the target offspring population corresponding to each of the multiple simulated planned driving trajectories.

[0240] The trajectory fitness evaluation result acquisition module is used to perform fitness evaluation processing on each target offspring population corresponding to each of the multiple simulated planned driving trajectories, and obtain the trajectory fitness evaluation results corresponding to each of the multiple target offspring populations. The trajectory fitness evaluation results indicate the degree of matching between the simulated planned driving trajectory corresponding to the target offspring population and the multiple simulated planned driving trajectories.

[0241] The optimal offspring population determination module is used to select the optimal offspring population from multiple target offspring populations based on the trajectory fitness evaluation results.

[0242] Parking control parameter set determination module: used to determine the optimal offspring population as the parking control parameter set.

[0243] In some embodiments, the apparatus further includes:

[0244] Fitness evaluation module: Used to evaluate the fitness of each individual in the initial population under multiple simulated driving trajectories, and obtain the evaluation value corresponding to each individual; the evaluation value indicates the degree of closeness between the individual and the optimal solution corresponding to the individual.

[0245] The first-generation population determination module is used to perform probability extraction on the initial population based on the evaluation value to obtain the first-generation population.

[0246] First generation population determination module: used to perform crossover and mutation processing on the first generation population to obtain the first generation population.

[0247] The second offspring population determination module is used to perform simulated annealing on the first offspring population based on a preset cooling rate to obtain the second offspring population.

[0248] The loop module is used to take the second offspring population as the initial population and sequentially execute the steps of fitness evaluation, probability extraction, crossover and mutation, and simulated annealing until the preset number of loops is reached.

[0249] The target offspring population acquisition module is also used to determine the population obtained when the number of loop executions reaches a preset number of loops as the target offspring populations corresponding to each of the multiple simulated planned driving trajectories.

[0250] In some embodiments, the apparatus further includes:

[0251] Crossover probability selection processing module: used to perform crossover probability selection processing on the first generation population to determine the crossover individuals that need to be crossovered in the first generation population.

[0252] Crossover processing module: Used to perform crossover processing on individuals to obtain the second generation population.

[0253] Mutation probability selection processing module: used to perform mutation probability selection processing on the second generation population to determine the mutant individuals that need to be mutated in the second generation population.

[0254] Mutation processing module: Used to process mutated individuals to obtain the first generation population.

[0255] In some embodiments, the apparatus further includes:

[0256] The fitness evaluation module is also used to evaluate the fitness of the first generation population and obtain the initial evaluation value corresponding to the first generation population.

[0257] The first acquisition module is used to acquire the initial temperature and the preset cooling rate; the initial temperature and the preset cooling rate are related to the individual's retention tolerance.

[0258] Individual Update Module: This module updates each gene in the first generation individuals of the first generation population with a preset step size, generating updated first generation individuals corresponding to the first generation population. The preset step size indicates whether to increase or decrease the parameters corresponding to the first generation individuals within a preset range.

[0259] The fitness evaluation module is also used to perform fitness evaluation on the updated first offspring individuals to obtain the update evaluation value corresponding to the updated first offspring individuals.

[0260] Evaluation value difference determination module: used to determine the evaluation value difference between the updated evaluation value and the initial evaluation value.

[0261] Individual evaluation probability acquisition module: used to perform evaluation calculations on the initial temperature, preset cooling rate and evaluation value difference to obtain the individual evaluation probability of any individual among the first generation individuals and the updated first generation individuals.

[0262] The loop module is also used to repeatedly execute the steps of update processing, fitness evaluation processing, difference processing and evaluation calculation processing with a preset step size until the preset number of loops is reached, and the second generation population is obtained based on the individual evaluation probability.

[0263] In some embodiments, the apparatus further includes:

[0264] The fitness evaluation module is also used to perform fitness evaluation processing on each target offspring population corresponding to each of the multiple simulated planned driving trajectories, to obtain the evaluation value under the multiple simulated planned driving trajectories for each target offspring population.

[0265] Average evaluation value acquisition module: used to average the multiple evaluation values ​​corresponding to each of the target offspring populations to obtain the average evaluation value corresponding to each target offspring population.

[0266] Evaluation value comparison module: used to compare and process the multiple average evaluation values ​​to obtain the trajectory fitness evaluation results corresponding to each of the multiple target offspring populations.

[0267] Filtering module: used to filter out the target offspring population with the largest average evaluation value from the trajectory fitness evaluation results.

[0268] Optimal offspring population determination module: used to determine the target offspring population with the largest average evaluation value as the optimal offspring population.

[0269] In some embodiments, the apparatus further includes:

[0270] Driving trajectory deviation information acquisition module: used to compare and process the current driving trajectory information and the planned driving trajectory information to obtain driving trajectory deviation information.

[0271] Vehicle acceleration information determination module: used to determine the vehicle acceleration information corresponding to the driving trajectory deviation information based on the longitudinal control model of the target vehicle.

[0272] Steering wheel angle information determination module: used to determine the driving trajectory deviation information and the steering wheel angle information of the vehicle corresponding to the steering angle ratio coefficient based on the lateral control model of the target vehicle.

[0273] The automatic parking control module 11 is also used to control the vehicle's driving trajectory during the automatic parking process based on acceleration information and steering wheel angle information.

[0274] Regarding the control device in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0275] Embodiments of this application also provide an electronic device, the device including a processor and a memory, the memory storing at least one instruction or at least one program segment, the at least one instruction or the at least one program segment being loaded and executed by the processor to implement the automatic parking method as described above.

[0276] Furthermore, Figure 12 A schematic diagram of the hardware structure of an electronic device for implementing the automatic parking method provided in the embodiments of this application is shown. The electronic device may participate in or include the apparatus provided in the embodiments of this application. Figure 12As shown, electronic device 1 may include one or more processors 902 (shown as 902a, 902b, ..., 902n in the figure) 902 (processor 902 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 904 for storing data, and a transmission device 906 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 12 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, electronic device 1 may also include... Figure 12 The more or fewer components shown, or having the same Figure 12 The different configurations shown.

[0277] It should be noted that the aforementioned one or more processors 902 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be wholly or partially embodied in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or wholly or partially integrated into any other element within the electronic device 1 (or mobile device). As involved in the embodiments of this application, the data processing circuit serves as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0278] The memory 904 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method in the embodiments of this application. The processor 902 executes various functional applications and data processing by running the software programs and modules stored in the memory 904, thereby realizing the above-mentioned automatic parking method. The memory 904 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 904 may further include memory remotely located relative to the processor 902, and these remote memories can be connected to the electronic device 1 via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0279] The transmission device 906 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the electronic device 1. In one example, the transmission device 906 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 906 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0280] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows a user to interact with the user interface of the electronic device 1 (or mobile device).

[0281] In this embodiment, the memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for the functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0282] Embodiments of this application also provide a computer storage medium storing at least one instruction or at least one program, which is loaded and executed by a processor to implement the automatic parking method described above.

[0283] Optionally, in this embodiment, the storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0284] In summary, the beneficial effects of the technical solution of this application are:

[0285] This application obtains optimized parking control parameters based on a preset optimization algorithm. Under diverse parking environments, the vehicle can be controlled to park using a single optimized parking control parameter, thus improving the efficiency of parking control parameter calibration. At the same time, this application considers the ratio between the front wheel angle and the rear wheel angle, and controls the vehicle's driving trajectory during automatic parking based on the angle ratio coefficient, driving trajectory information, and planned driving trajectory information, thereby improving the safety, stability, practicality, and comfort of automatic parking.

[0286] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0287] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and apparatus embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0288] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0289] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An automatic parking method characterized by, The method is applied to a four-wheel steering vehicle, and comprises the following steps: In response to a vehicle parking operation, a set of parking control parameters is obtained, the set of parking control parameters being obtained by optimizing and calibrating a reference set of parking control parameters based on a preset optimization algorithm, the reference set of parking control parameters being obtained by calibrating and extracting parameters based on a parameter range of a plurality of parking control parameters; An automatic parking of the vehicle is controlled based on the parking control parameters in the set of parking control parameters; In the automatic parking process of the vehicle, a current front wheel steering angle of the vehicle, current trajectory information and planned trajectory information of the vehicle are obtained, the planned trajectory information representing a planned trajectory of the vehicle on a target parking section between a target parking start point and a target parking end point, and the current trajectory information representing actual trajectory information of the vehicle formed in the parking process; A front-rear wheel proportionality coefficient corresponding to the current front wheel steering angle is determined, the front-rear wheel proportionality coefficient indicating a ratio between a rear wheel steering angle and the front wheel steering angle of the vehicle; The trajectory of the vehicle in the automatic parking process is controlled based on the front-rear wheel proportionality coefficient, the current trajectory information and the planned trajectory information.

2. The automatic parking method according to claim 1, characterized by, The obtaining of the set of parking control parameters comprises the following steps: A plurality of simulated planned trajectories are obtained; The reference set of parking control parameters is set as an initial population of the preset optimization algorithm, and the control parameters in the set of parking control parameters are one-to-one corresponding to genes in individuals of the initial population; Optimization processing of each individual of the initial population is performed based on the preset optimization algorithm to approach an optimal solution, so as to obtain a target offspring population corresponding to each of the plurality of simulated planned trajectories; For each of the target offspring populations corresponding to each of the plurality of simulated planned trajectories, fitness evaluation processing corresponding to each of the plurality of simulated planned trajectories is performed, so as to obtain a trajectory fitness evaluation result corresponding to each of the plurality of target offspring populations, the trajectory fitness evaluation result indicating a matching degree between the simulated planned trajectory corresponding to the target offspring population and the plurality of simulated planned trajectories; An optimal offspring population is selected from the plurality of target offspring populations based on the trajectory fitness evaluation result; The optimal offspring population is determined as the set of parking control parameters.

3. The automatic parking method according to claim 2, characterized by, The optimization processing of each individual of the initial population based on the preset optimization algorithm to approach an optimal solution, so as to obtain a target offspring population corresponding to each of the plurality of simulated planned trajectories, comprises the following steps: Under the plurality of simulated planned trajectories, fitness evaluation processing of each individual of the initial population is performed to obtain an evaluation value corresponding to each individual, the evaluation value indicating an approaching degree between the individual and an optimal solution corresponding to the individual; Probability extraction processing of the initial population is performed based on the evaluation value to obtain a first generation population; Crossing and mutation processing of the first generation population is performed to obtain a first offspring population; Simulated annealing processing of the first offspring population is performed based on a preset cooling rate to obtain a second offspring population; The second sub-population is taken as the initial population, and the steps of the fitness evaluation process, the probability extraction process, the crossover and mutation process, and the simulated annealing process are sequentially performed in cycles until the number of cycles reaches a preset number of cycles; The population obtained when the number of cycles reaches the preset number of cycles is determined as the target sub-population corresponding to each of the plurality of simulated planning driving trajectories.

4. The automatic parking method according to claim 3, characterized by, The crossover and mutation process on the first generation population to obtain a first sub-population includes: A crossover probability selection process is performed on the first generation population to determine crossover individuals in the first generation population that need to be crossed; A crossover process is performed on the crossover individuals to obtain a second generation population; A mutation probability selection process is performed on the second generation population to determine mutation individuals in the second generation population that need to be mutated; A mutation process is performed on the mutation individuals to obtain a first sub-population.

5. The automatic parking method according to claim 3, characterized by, The simulated annealing process on the first sub-population based on a preset cooling rate to obtain a second sub-population includes: A fitness evaluation process is performed on the first sub-population to obtain an initial evaluation value corresponding to the first sub-population; An initial temperature and a preset cooling rate are obtained; the initial temperature and the preset cooling rate are associated with the retention tolerance of an individual; A preset step update process is performed on each gene in the first sub-individuals of the first sub-population to generate updated first sub-individuals corresponding to the first sub-population; the preset step indicates that the parameters corresponding to each gene in the first sub-individuals are increased or decreased by a preset value range; A fitness evaluation process is performed on the updated first sub-individuals to obtain an updated evaluation value corresponding to the updated first sub-individuals; An evaluation value difference between the updated evaluation value and the initial evaluation value is determined; An evaluation calculation process is performed on the initial temperature, the preset cooling rate, and the evaluation value difference to obtain an individual evaluation probability of any individual of the first sub-individuals and the updated first sub-individuals; The steps of the preset step update process, the fitness evaluation process, the difference process, and the evaluation calculation process are performed in cycles until the number of cycles reaches a preset number of cycles, and the second sub-population is obtained based on the individual evaluation probability.

6. The automatic parking method according to claim 2, characterized by, The fitness evaluation process corresponding to each of the plurality of simulated planning driving trajectories is performed on each of the target sub-populations corresponding to the plurality of simulated planning driving trajectories to obtain a trajectory fitness evaluation result corresponding to each of the plurality of target sub-populations includes: The fitness evaluation process corresponding to each of the plurality of simulated planning driving trajectories is performed on each of the target sub-populations corresponding to the plurality of simulated planning driving trajectories to obtain an evaluation value of the plurality of simulated planning driving trajectories corresponding to each of the target sub-populations; An average value is obtained by averaging the plurality of evaluation values corresponding to each of the target sub-populations, respectively, to obtain an average evaluation value corresponding to each of the target sub-populations; The multiple average evaluation values are compared to obtain trajectory fitness evaluation results corresponding to the multiple target sub-populations respectively; The optimal sub-population is selected from the multiple target sub-populations based on the trajectory fitness evaluation results, and the method comprises the steps of: Selecting a target sub-population with the maximum average evaluation value from the trajectory fitness evaluation results; The target sub-population with the maximum average evaluation value is determined as the optimal sub-population.

7. The automatic parking method according to any one of claims 1 to 6, characterized in that, The driving trajectory of the vehicle in the automatic parking process is controlled based on the steering angle proportion coefficient, the current driving trajectory information and the planned driving trajectory information, and the method comprises the steps of: Comparing the current driving trajectory information and the planned driving trajectory information to obtain driving trajectory deviation information; Determining vehicle acceleration information corresponding to the driving trajectory deviation information based on a longitudinal control model of the target vehicle; Determining steering wheel steering angle information of the vehicle corresponding to the driving trajectory deviation information and the steering angle proportion coefficient based on a lateral control model of the target vehicle; Controlling the driving trajectory of the vehicle in the automatic parking process based on the acceleration information and the steering wheel steering angle information.

8. A control device for automatic parking, characterized by comprising: The device comprises: A parking control parameter set acquisition module is configured to acquire a parking control parameter set in response to a vehicle parking operation, wherein the parking control parameter set is obtained by optimizing and calibrating a reference parking control parameter set based on a preset optimization algorithm, and the reference parking control parameter set is obtained by calibrating and extracting parameters based on a parameter range of multiple parking control parameters; An automatic parking control module is configured to control the vehicle to automatically park based on the parking control parameters in the parking control parameter set; An information acquisition module is configured to acquire current front wheel steering angle, current driving trajectory information and planned driving trajectory information of the vehicle during automatic parking of the vehicle, wherein the planned driving trajectory information represents a planned driving trajectory of the vehicle on a target parking section between a target parking starting point and a target parking ending point, and the current driving trajectory information represents actual trajectory information formed by the vehicle during parking; A front-rear wheel proportion coefficient acquisition module is configured to determine a front-rear wheel proportion coefficient corresponding to the current front wheel steering angle, wherein the front-rear wheel proportion coefficient indicates a ratio between rear wheel steering angle and front wheel steering angle of the vehicle; The automatic parking control module is further configured to control the driving trajectory of the vehicle in the automatic parking process based on the steering angle proportion coefficient, the current driving trajectory information and the planned driving trajectory information.

9. An electronic device, comprising: The device comprises a processor and a memory, and the memory stores at least one instruction or at least one program, which is loaded and executed by the processor to implement the automatic parking method according to any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores at least one instruction or at least one program, which is loaded and executed by the processor to implement the automatic parking method according to any one of claims 1-7.

Images