Parking feasible region calculation method and device, vehicle and storage medium

By using genetic algorithms and random roulette strategies in automatic parking systems, the problem of low accuracy of parking feasible domains is solved, the accuracy and reliability of parking functions are improved, and the Accidents and waste of resources.

CN119078800APending Publication Date: 2024-12-06BEIJING ELECTRIC VEHICLE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411137655.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The calculation accuracy of the parking feasible domain during automatic parking is low, resulting in poor actual effect of the parking function, increasing accident risk and resource waste.

Method used

By obtaining vehicle dynamics models, parking conditions and parking space restrictions, establish constraints for the initial parking feasible domain, and solve the nonlinear optimization function based on the genetic algorithm to calculate the final parking feasible domain, using the preset genetic strategy and The iterative process of random roulette strategy optimization and improve the accuracy of the feasible parking domain.

Benefits of technology

It greatly improves the accuracy and reliability of automatic parking, reduces parking misjudgment, and enhances the reliability and safety of parking functions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119078800A_ABST
    Figure CN119078800A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of vehicles, in particular to a parking feasible region calculation method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining a vehicle dynamic model, a parking condition, a parking space limiting condition, a parking space position and an initial parking feasible region of a current vehicle, establishing a constraint condition of an initial parking feasible region according to the vehicle dynamics model, the parking condition and the parking space limiting condition, establishing a path planning function according to the vehicle dynamics model and the parking space position, obtaining a nonlinear optimization function according to the constraint condition and the path planning function, and solving the nonlinear optimization function based on a preset genetic strategy, and obtaining an initial iteration point, carrying out iterative solution on the path planning function by using the initial iteration point, and calculating a final parking feasible region. Therefore, the problem that the calculation accuracy of the parking feasible region is low in the automatic parking process is solved, the parking feasible region is judged and optimized, and the accuracy and reliability of automatic parking are greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a method, device, vehicle and storage medium for calculating a parking feasible domain. Background Art

[0002] With the rapid development of my country's automobile industry, the production and sales volume of automobiles has reached nearly 30 million. More and more cars have entered thousands of households, facilitating the needs of the general public. However, the resulting parking problem has become increasingly serious. The narrow parking space and limited field of vision have made parking more difficult for drivers, which in turn has led to the continuous occurrence and increase of accidents in manual parking. Therefore, the development of automatic parking technology is solving people's pain points about parking difficulties.

[0003] Automatic parking technology is a new type of automatic parking technology that replaces manual operation by the driver. It mainly includes technical modules such as parking space recognition, path planning and motion control. Among them, the determination of the feasible domain of parking (driving area where parking function can be performed) is an important part of performing the parking function. This is the basis for the automatic parking function technology to activate the parking function after the vehicle finds a parking space.

[0004] Genetic algorithm originated in the 1970s. The algorithm was designed and proposed based on the evolutionary laws of organisms in nature. It is a computational model of biological evolution that simulates the natural selection and genetic mechanisms of Darwin's theory of biological evolution. It is a method of searching for the optimal solution by simulating the natural evolution process. The algorithm uses mathematical methods and computer simulation operations to convert the problem-solving process into a process similar to the crossover and mutation of chromosome genes in biological evolution. When solving more complex combinatorial optimization problems, it can usually obtain better optimization results faster than some conventional optimization algorithms.

[0005] In the related technology, the vehicle's dynamic model is calculated to establish a vehicle kinematic model function, which is generally a nonlinear function. The collision constraints are established during the vehicle's reversing into a parking space by simulating the vehicle's dynamic model. A nonlinear optimization function with constraints is abstracted based on the constraints and the vehicle kinematic model function, and finally the feasible parking domain that meets the vehicle dynamic model and parking constraints is calculated.

[0006] However, the method of calculating the parking feasible domain involves nonlinear optimization of functions, and its essence is an iterative function optimization process. The relevant technology adopts the method of randomly selecting initial points for initial iteration, which will result in poor convergence effect and low accuracy of the parking feasible domain results, thereby affecting the actual effect of the parking function. Summary of the invention

[0007] The present application provides a parking feasible domain calculation method, device, electronic device and storage medium to solve the problem of low calculation accuracy of the parking feasible domain during automatic parking, judge and optimize the parking feasible domain, and greatly improve the accuracy and reliability of automatic parking.

[0008] A first aspect of the present application provides a method for calculating a parking feasible domain, comprising the following steps: obtaining a vehicle dynamics model, parking conditions, parking space restrictions, parking space positions and an initial parking feasible domain of a current vehicle; establishing constraints of the initial parking feasible domain according to the vehicle dynamics model, the parking conditions and the parking space restrictions, and establishing a path planning function according to the vehicle dynamics model and the parking space positions; obtaining a nonlinear optimization function according to the constraints and the path planning function, solving the nonlinear optimization function based on a preset genetic strategy to obtain an initial iteration point, and iteratively solving the path planning function using the initial iteration point to calculate a final parking feasible domain.

[0009] Optionally, in some embodiments, after calculating the final parking feasible domain, the method further includes: predicting the parking of the current vehicle based on the final parking feasible domain, so as to complete the parking action of the current vehicle according to the prediction result.

[0010] Optionally, in some embodiments, the preset genetic strategy is: obtaining multiple initial target parking feasible domains; generating multiple initial target parking feasible domain genetic operator populations, and calculating the fitness of each initial target parking feasible domain genetic operator, and judging whether the preset iteration condition restriction is satisfied based on the fitness; if the preset iteration condition restriction is not satisfied, selecting two parking feasible domain calculation genetic operator populations based on a preset random roulette strategy, and optimizing the two parking feasible domain calculation genetic operator populations, and re-judging whether the fitness satisfies the preset iteration condition restriction, until the fitness satisfies the preset iteration condition restriction, and obtaining the optimal parking feasible domain solution.

[0011] Optionally, in some embodiments, the optimizing processing of calculating the genetic operator populations of the two parking feasible domains includes: performing a copy operation on any one of the two parking feasible domains, and adding the copy result to a new population; performing a mutation operation on the remaining parking feasible domain of the two parking feasible domains, and adding the mutation result to the new population; performing a crossover operation on the two parking feasible domains, and adding the crossover result to the new population.

[0012] Optionally, in some embodiments, the preset random roulette strategy is: determining a genetic operator set point, wherein the set point belongs to a set of genetic operator points that meet the parking feasible domain constraint condition;

[0013] Based on the genetic operator set points and a preset random number, a set of parking genetic operators whose selection probability is greater than or equal to the preset random number is screened out from the set of genetic operator points that meet the parking feasible domain constraint conditions, wherein the preset random number obeys a uniform distribution and is within an interval formed by zero and a function fitness probability calculated by the parking feasible domain genetic operator; based on the set of parking genetic operators, a parking feasible domain genetic operator calculation point is screened out using a target roulette algorithm.

[0014] A second aspect of the present application provides a parking feasible domain calculation device, including: an acquisition module, used to acquire a vehicle dynamics model, parking conditions, parking space restrictions, parking space positions and an initial parking feasible domain of a current vehicle; a modeling module, used to establish constraints of the initial parking feasible domain according to the vehicle dynamics model, the parking conditions and the parking space restrictions, and to establish a path planning function according to the vehicle dynamics model and the parking space positions; a calculation module, used to obtain a nonlinear optimization function according to the constraints and the path planning function, and to solve the nonlinear optimization function based on a preset genetic strategy to obtain an initial iteration point, and to iteratively solve the path planning function using the initial iteration point to calculate a final parking feasible domain.

[0015] Optionally, in some embodiments, after calculating the final parking feasible domain, the calculation module is further used to: predict the parking of the current vehicle based on the final parking feasible domain, so as to complete the parking action of the current vehicle according to the prediction result.

[0016] Optionally, in some embodiments, the preset genetic strategy is: obtaining multiple initial target parking feasible domains; generating multiple initial target parking feasible domain genetic operator populations, and calculating the fitness of each initial target parking feasible domain genetic operator, and judging whether the preset iteration condition restriction is satisfied based on the fitness; if the preset iteration condition restriction is not satisfied, selecting two parking feasible domain calculation genetic operator populations based on a preset random roulette strategy, and optimizing the two parking feasible domain calculation genetic operator populations, and re-judging whether the fitness satisfies the preset iteration condition restriction, until the fitness satisfies the preset iteration condition restriction, and obtaining the optimal parking feasible domain solution.

[0017] Optionally, in some embodiments, the optimizing processing of calculating the genetic operator populations of the two parking feasible domains includes: performing a copy operation on any one of the two parking feasible domains, and adding the copy result to a new population; performing a mutation operation on the remaining parking feasible domain of the two parking feasible domains, and adding the mutation result to the new population; performing a crossover operation on the two parking feasible domains, and adding the crossover result to the new population.

[0018] Optionally, in some embodiments, the preset random roulette strategy is: determining a set of genetic operator points, wherein the set points belong to a set of genetic operator points that meet the parking feasible domain constraint conditions; based on the genetic operator set points and a preset random number, screening out a set of parking genetic operators whose selection probability is greater than or equal to the preset random number from the set of genetic operator points that meet the parking feasible domain constraint conditions, wherein the preset random number obeys a uniform distribution and the preset random number is within an interval between zero and the function fitness probability calculated by the parking feasible domain genetic operator; based on the set of parking genetic operators, screening out parking feasible domain genetic operator calculation points using a target roulette algorithm.

[0019] A third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the parking feasible domain calculation method as described in the above embodiment.

[0020] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the parking feasible domain calculation method as described in the above embodiment.

[0021] Therefore, the present application establishes the constraints of the initial parking feasible domain according to the vehicle dynamics model, parking conditions and parking space restrictions, and establishes a path planning function according to the vehicle dynamics model and parking space position, and obtains a nonlinear optimization function by combining the constraints and the path planning function, and solves the nonlinear optimization function based on a genetic algorithm to obtain an initial iteration point, and uses the initial iteration point to iteratively solve the path planning function to calculate the final parking feasible domain. Therefore, the problem of low calculation accuracy of the parking feasible domain during automatic parking is solved, and the parking feasible domain is judged and optimized, which greatly improves the accuracy and reliability of automatic parking.

[0022] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0024] Figure 1 A schematic diagram of a parking collision provided according to a specific embodiment of the present application;

[0025] Figure 2 A schematic diagram of a feasible parking area provided according to a specific embodiment of the present application;

[0026] Figure 3 A flowchart of a method for calculating a parking feasible region according to an embodiment of the present application;

[0027] Figure 4 A schematic diagram of a random roulette algorithm process provided according to a specific embodiment of the present application;

[0028] Figure 5 A schematic diagram of an improved genetic algorithm process provided according to a specific embodiment of the present application;

[0029] Figure 6 A schematic diagram of a parking feasible region calculation method process provided according to a specific embodiment of the present application;

[0030] Figure 7 A schematic diagram of a parking feasible region calculation device provided according to an embodiment of the present application;

[0031] Figure 8 It is a schematic diagram of the structure of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0033] The following describes the parking feasible domain calculation method, device, vehicle and storage medium of the embodiment of the present application with reference to the accompanying drawings. In view of the problem of low calculation accuracy of the parking feasible domain in the automatic parking process mentioned in the above background technology, the present application provides a parking feasible domain calculation method, in which the vehicle dynamics model, parking conditions, parking space restrictions, parking space position and initial parking feasible domain of the current vehicle are obtained, and the constraints of the initial parking feasible domain are established according to the vehicle dynamics model, parking conditions and parking space restrictions, and the path planning function is established according to the vehicle dynamics model and the parking space position, and the nonlinear optimization function is obtained according to the constraints and the path planning function, and the nonlinear optimization function is solved based on the preset genetic strategy to obtain the initial iteration point, and the path planning function is iteratively solved using the initial iteration point to calculate the final parking feasible domain. In this way, the problem of low calculation accuracy of the parking feasible domain in the automatic parking process is solved, the parking feasible domain is judged and optimized, and the accuracy and reliability of automatic parking are greatly improved.

[0034] Before introducing the embodiments of the present application, Figure 1 and Figure 2 , introduce the calculation method of the parking feasible domain in the related technology.

[0035] Specifically, Figure 1 A schematic diagram of a parking collision provided in an embodiment of the present application, Figure 2 The schematic diagram of the parking feasible region provided in the embodiment of the present application, the method for calculating the feasible region mainly includes the following steps:

[0036] 1. Calculate the vehicle's dynamic model and establish the vehicle's kinematic model function, which is generally a nonlinear function;

[0037] 2. If Figure 1 As shown, collision constraints are established during the process of simulating the vehicle reversing into a garage according to the vehicle dynamics model;

[0038] 3. If Figure 2 As shown, according to step 1 and step 2, a nonlinear optimization function with constraints is abstracted, and finally the parking feasible domain that meets the vehicle dynamics model and parking constraints is calculated.

[0039] according to Figure 1 and Figure 2 Solve the following formula, where f(x) is based on Figure 2 Parking feasible domain function for solving parking trajectory, g i (x) and h j (x) is based on Figure 1 Parking constraints established.

[0040]

[0041] Where x=(x1,x2,...,x n ) T , which represents the feasible domain solution of problem (1-1).

[0042] However, the method for calculating the parking feasible region has poor convergence effect, and the accuracy of the parking feasible region result is low. If the calculation of the parking feasible region is inaccurate, it will cause misjudgment to the automatic parking system. For example, if the area where parking is not allowed is misjudged as an area where parking is allowed, it may cause a parking accident; if the area where parking is allowed is misjudged as an area where parking is not allowed, it will cause unnecessary waste of personal and public resources. Therefore, in order to solve the above problems, this application proposes a nonlinear optimization parking feasible region method based on genetic algorithm.

[0043] Specifically, Figure 3 A flowchart of a method for calculating a parking feasible region provided in an embodiment of the present application.

[0044] like Figure 3 As shown, the parking feasible region calculation method includes the following steps:

[0045] In step S301, a vehicle dynamics model, parking conditions, parking space restriction conditions, parking space position and an initial parking feasible region of the current vehicle are obtained.

[0046] Among them, the vehicle dynamics model of the current vehicle is obtained by calculation during the actual vehicle control process, when the vehicle is in a scene such as turning. The parking conditions of the embodiment of the present application are the conditions for automatic parking of the vehicle, which may be the prerequisites for automatic parking such as "supporting lateral and longitudinal automatic parking" and "fully automatic, without the need to control brakes and gears". The parking space restriction conditions of the embodiment of the present application are used as parameters for calculating the parking position to regulate parking, such as the distance between the parking space and the non-motor vehicle parking space and other conditions that restrict parking. The parking feasible domain calculation method of the embodiment of the present application also needs to obtain the parking space position for planning the direction of automatic parking. In addition, the method for obtaining the initial parking feasible domain of the embodiment of the present application is similar to the method described in the above-mentioned related technology. In order to avoid redundancy, it will not be repeated here.

[0047] It can be understood that the automatic parking technology should be divided into four steps in terms of implementation steps, namely, parking space identification, parking feasible domain judgment, path planning, and motion control for automatic parking. Therefore, this application needs to obtain parameters such as parking conditions, parking space restrictions, and parking space positions for application in parking space identification and path planning. In addition, parking feasible domain judgment is a prerequisite for path planning and motion control. The vehicle dynamics model, parking conditions, parking space restrictions, parking space positions, and initial parking feasible domain of the current vehicle obtained through step S301 in the embodiment of the present application provide a basis for the subsequent calculation and optimization of the parking feasible domain, and this application does not make specific limitations on the above data.

[0048] In step S302, the constraints of the initial parking feasible domain are established according to the vehicle dynamics model, parking conditions and parking space restriction conditions, and a path planning function is established according to the vehicle dynamics model and parking space position.

[0049] It is understandable that after the current vehicle detects the target parking space, in order to ensure successful parking, the vehicle needs to stay at a suitable starting position, and then control the vehicle to reverse into the space. Therefore, it is necessary to calculate the starting area that meets the parking conditions. Then, for a target parking space, the embodiment of the present application takes into account the constraints of the parking space boundary and the road boundary to establish the constraints of the initial parking feasible domain. In addition, the embodiment of the present application establishes a path planning function based on the vehicle dynamics model and the parking space position, and the path planning function is a nonlinear function.

[0050] It should be noted that the constraint conditions of the initial parking feasible domain and the calculation method of the path planning function in the embodiment of the present application are similar to the methods described in the above-mentioned related technologies. To avoid redundancy, they are not described here.

[0051] In step S303, a nonlinear optimization function is obtained according to the constraint conditions and the path planning function, and the nonlinear optimization function is solved based on a preset genetic strategy to obtain an initial iteration point, and the path planning function is iteratively solved using the initial iteration point to calculate the final parking feasible domain.

[0052] It is understandable that the nonlinear optimization function obtained according to the constraint conditions and the path planning function is used to calculate the parking feasible domain, which is essentially an iterative function optimization process, and the method of randomly selecting the initial point will result in poor convergence effect and low accuracy of the parking feasible domain result. Therefore, the present application solves this defect by solving the nonlinear optimization function according to the preset genetic strategy to obtain the initial iteration point.

[0053] Among them, the parking feasible domain calculation combined with the vehicle dynamics model in this application can be abstracted as a method for solving the extreme value of a function. A nonlinear optimization function is obtained according to the constraints and path planning function obtained in step S302. This function is not linear, so it is also called nonlinear function extreme value seeking.

[0054] Specifically, according to the constraints and the path planning function, a nonlinear optimization function with Lagrangian parameters is transformed, and the nonlinear optimization function is evaluated using an improved genetic algorithm to solve the extreme point, which is the initial iteration point of the function for solving the parking constraint. Furthermore, according to the solved initial iteration point, the application uses other optimization functions such as gradient descent or Newton iteration to iteratively solve the path planning curve, and finally calculates the feasible domain of parking.

[0055] Optionally, in some embodiments, the preset genetic strategy is: obtaining multiple initial target parking feasible domains; generating multiple initial target parking feasible domain genetic operator populations, and calculating the fitness of each initial target parking feasible domain genetic operator, and judging whether the preset iteration condition limit is satisfied based on the fitness; if the preset iteration condition limit is not satisfied, selecting two parking feasible domain calculation genetic operator populations based on the preset random roulette strategy, and optimizing the two parking feasible domain calculation genetic operator populations, and re-judging whether the fitness satisfies the preset iteration condition limit, until the fitness satisfies the preset iteration condition limit, and the optimal parking feasible domain solution is obtained.

[0056] It should be noted that the preset genetic strategy of the embodiment of the present application is an improvement on the genetic algorithm in the above-mentioned related technology, and its purpose is to use the algorithm to obtain a more accurate parking feasible domain when solving the parking feasible domain. The genetic algorithm of the embodiment of the present application needs to obtain the initial iterative parking feasible domain genetic operator population, calculate the fitness of each parking feasible domain genetic operator, and then the embodiment of the present application sets a preset iteration condition restriction, and its purpose is to determine whether the fitness of each parking feasible domain genetic operator meets the iteration condition, that is, if the fitness of the parking feasible domain genetic operator meets the restriction of the iteration condition, then the parking feasible domain genetic operator population is designated as the parking feasible domain solution, and the process ends.

[0057] In some cases, the fitness of the parking feasible domain genetic operator does not meet the restrictions of the iteration conditions. Therefore, the embodiment of the present application proposes a preset random roulette strategy to screen the genetic operator population that does not meet the restrictions of the iteration conditions, and the screened genetic operator population is processed accordingly through an improved genetic algorithm until the optimal parking feasible domain solution is screened out.

[0058] Optionally, in some embodiments, the preset random roulette strategy is: determining a set of genetic operator points, wherein the set points belong to a set of genetic operator points that meet the parking feasible domain constraint conditions; based on the genetic operator set points and a preset random number, screening out a set of parking genetic operators whose selection probability is greater than or equal to a preset random number from the set of genetic operator points that meet the parking feasible domain constraint conditions, wherein the preset random number obeys a uniform distribution and the preset random number is within an interval between zero and the function fitness probability calculated by the parking feasible domain genetic operator; based on the set of parking genetic operators, screening out parking feasible domain genetic operator calculation points using a target roulette algorithm.

[0059] It should be noted that the core idea of ​​the roulette strategy is to allocate corresponding selection probabilities according to the possibilities of different options, ensuring that the high-probability options have a high probability of being selected while retaining the possibility of other options with relatively low probabilities being selected. The traditional roulette algorithm can avoid the local optimal trap, however, its disadvantage is that the convergence speed is slow. Therefore, the preset random roulette strategy proposed in this application is an improvement on the traditional roulette strategy, and this improved roulette strategy has a faster convergence speed.

[0060] Specifically, Figure 4 A schematic diagram of the random roulette algorithm process provided in the embodiment of the present application is shown in FIG. Figure 4 As shown, the random roulette algorithm of the embodiment of the present application includes the following steps:

[0061] (1) First, determine the genetic operator set point k, k∈inital-able, where inital-able is the set of genetic operator points that meet the parking feasible domain constraints;

[0062] (2) Then, by adding a random number R∈[0, max(P k )) for comparison, where P k The function fitness probability calculated for the parking feasible domain genetic operator is obtained, and the set of parking genetic operators with a selection probability P greater than or equal to R is screened out;

[0063] (3) Finally, the traditional roulette algorithm is used to select the genetic operator calculation points in the parking feasible domain.

[0064] Therefore, the random roulette algorithm of the present application is actually an improved algorithm based on the traditional roulette algorithm and the greedy algorithm, which can not only improve the convergence speed of the traditional roulette algorithm, but also avoid the defect of the greedy algorithm that is easy to fall into the local optimal value. When the random number R = 0, the algorithm is equivalent to the traditional roulette algorithm and can avoid the local optimal value trap; when R = max(P k ), the algorithm is equivalent to a greedy algorithm and has a faster convergence speed.

[0065] Optionally, in some embodiments, the genetic operator populations calculated for the two parking feasible domains are optimized, including: copying any one of the two parking feasible domains and adding the copy result to a new population; mutating the remaining parking feasible domains of the two parking feasible domains and adding the mutated result to the new population; and crossovering the two parking feasible domains and adding the crossover result to the new population.

[0066] It is understandable that after the feasible domain genetic operator population that does not meet the preset iteration condition restrictions is screened out based on the random roulette strategy, the feasible domain genetic operator calculation point needs to be optimized and the fitness is re-determined whether it meets the iteration condition restrictions until the fitness meets the preset iteration condition restrictions to obtain the optimal parking feasible domain solution. In addition, the present application does not specifically limit the iteration method, and those skilled in the art can select the iteration method as needed, for example, Newton iteration or other nonlinear optimization algorithms can be used.

[0067] Specifically, the embodiment of the present application is based on a genetic algorithm, which converts the process of optimizing the calculation points of the feasible domain genetic operator into a process similar to the crossover and mutation of chromosome genes in biological evolution, that is, selecting a parking feasible domain calculation genetic operator and copying it to a new population; crossing two parking feasible domain calculation genetic operators, and adding the crossover result to the new population; selecting a parking feasible domain calculation genetic operator to perform a mutation operation, and then inserting the mutation result into the new population.

[0068] Therefore, the initial parking feasible domain genetic operator population is first screened out by calculating its fitness, and the genetic operators that do not meet the iteration conditions are selected using a random roulette algorithm. The genetic operator population is then optimized through selection, crossover, and mutation operations, and is further iterated until the optimal parking feasible domain solution is selected.

[0069] Furthermore, in order to facilitate those skilled in the art to further understand the steps of the improved genetic algorithm in the embodiment of the present application, they are described in detail below with reference to schematic diagrams and specific embodiments.

[0070] Specifically, Figure 5 As shown, the improved genetic algorithm includes the following steps:

[0071] 1) Generate the initial iterative parking feasible domain genetic operator population;

[0072] 2) Calculate the fitness of each parking feasible region genetic operator and determine whether the iteration condition is met. If the iteration condition is met, end the process, otherwise proceed to the next step;

[0073] 3) Using random roulette algorithm to screen the genetic operator population of parking feasible domain;

[0074] 4) Select, crossover, and mutate the population selected in step 3;

[0075] 5) Proceed to the next step iteration until the optimal parking feasible domain solution is screened out.

[0076] Therefore, according to the improved genetic algorithm proposed in the embodiment of the present application, the parking feasible domain genetic operator population is calculated, judged, and optimized. The present application can obtain a more accurate parking feasible domain solution, thereby achieving accurate judgment of the automatic parking function.

[0077] Furthermore, the parking feasible region calculation method proposed in the present application is described in detail below through an embodiment.

[0078] Specifically, Figure 6 A schematic diagram showing the process of the parking feasible domain calculation method provided in the embodiment of the present application is shown as follows: Figure 6 As shown, the parking feasible region calculation method includes the following steps:

[0079] 1. Establish the constraints of the parking feasible domain based on the vehicle dynamics model, parking conditions, parking space restrictions, etc.;

[0080] 2. Establish the curve function of path planning based on the vehicle dynamics model and parking space location;

[0081] 3. According to the path planning function and constraint conditions established in step 1 and step 2, a nonlinear optimization function with Lagrangian parameters is transformed, and the nonlinear optimization function is evaluated using an improved genetic algorithm to solve the extreme value point, which is the initial iteration point of the function for solving the parking constraint condition;

[0082] 4. According to the initial iteration point solved in step 3, the path planning curve is iteratively solved using other optimization functions such as gradient descent or Newton iteration, and finally the feasible domain of parking is calculated.

[0083] Therefore, the present application establishes a vehicle dynamics model and parking constraints, and then establishes a parking trajectory function, solves the initial iterative parking feasible domain based on an improved genetic algorithm, and iteratively optimizes the initial iterative parking feasible domain to solve the final accurate parking feasible domain.

[0084] Optionally, in some embodiments, after calculating the final parking feasible domain, the method further includes: predicting the parking of the current vehicle based on the final parking feasible domain, so as to complete the parking action of the current vehicle according to the prediction result.

[0085] Specifically, after calculating the final parking feasible domain, the current vehicle predicts the parking action of the current vehicle based on the final parking feasible domain, the target parking space and the surrounding environment parameters. If the vehicle parking system determines that the current vehicle meets the stopping criteria, the current vehicle is controlled to complete the parking action.

[0086] According to the parking feasible domain calculation method proposed in the embodiment of the present application, the constraints of the initial parking feasible domain are established based on the vehicle dynamics model, parking conditions and parking space restrictions, and the path planning function is established according to the vehicle dynamics model and the parking space position. The nonlinear optimization function is obtained by combining the constraints and the path planning function, and the nonlinear optimization function is solved based on the genetic algorithm to obtain the initial iteration point, and the path planning function is iteratively solved using the initial iteration point to calculate the final parking feasible domain. In this way, the problem of low calculation accuracy of the parking feasible domain in the automatic parking process is solved, the parking feasible domain is judged and optimized, and the accuracy and reliability of automatic parking are greatly improved.

[0087] Next, the parking feasible region calculation device proposed in accordance with the embodiment of the present application is described with reference to the accompanying drawings.

[0088] Figure 7 4 is a block diagram of a parking feasible domain calculation device according to an embodiment of the present application.

[0089] like Figure 7 As shown, the parking feasible region calculation device 10 includes: an acquisition module 100 , a modeling module 200 and a calculation module 300 .

[0090] Among them, the acquisition module 100 is used to obtain the vehicle dynamics model, parking conditions, parking space restrictions, parking space position and initial parking feasible domain of the current vehicle; the modeling module 200 is used to establish the constraints of the initial parking feasible domain according to the vehicle dynamics model, parking conditions and parking space restrictions, and establish a path planning function according to the vehicle dynamics model and the parking space position; the calculation module 300 is used to obtain a nonlinear optimization function according to the constraints and the path planning function, and solve the nonlinear optimization function based on a preset genetic strategy to obtain an initial iteration point, and use the initial iteration point to iteratively solve the path planning function to calculate the final parking feasible domain.

[0091] Optionally, in some embodiments, after calculating the final parking feasible domain, the calculation module 300 is further used to: predict the parking of the current vehicle based on the final parking feasible domain, so as to complete the parking action of the current vehicle according to the prediction result.

[0092] Optionally, in some embodiments, the preset genetic strategy is: obtaining multiple initial target parking feasible domains; generating multiple initial target parking feasible domain genetic operator populations, and calculating the fitness of each initial target parking feasible domain genetic operator, and judging whether the preset iteration condition limit is satisfied based on the fitness; if the preset iteration condition limit is not satisfied, selecting two parking feasible domain calculation genetic operator populations based on the preset random roulette strategy, and optimizing the two parking feasible domain calculation genetic operator populations, and re-judging whether the fitness satisfies the preset iteration condition limit, until the fitness satisfies the preset iteration condition limit, and the optimal parking feasible domain solution is obtained.

[0093] Optionally, in some embodiments, the genetic operator populations calculated for the two parking feasible domains are optimized, including: copying any one of the two parking feasible domains and adding the copy result to a new population; mutating the remaining parking feasible domains of the two parking feasible domains and adding the mutated result to the new population; and crossovering the two parking feasible domains and adding the crossover result to the new population.

[0094] Optionally, in some embodiments, the preset random roulette strategy is: determining a set of genetic operator points, wherein the set points belong to a set of genetic operator points that meet the parking feasible domain constraint conditions; based on the genetic operator set points and a preset random number, screening out a set of parking genetic operators whose selection probability is greater than or equal to a preset random number from the set of genetic operator points that meet the parking feasible domain constraint conditions, wherein the preset random number obeys a uniform distribution and the preset random number is within an interval between zero and the function fitness probability calculated by the parking feasible domain genetic operator; based on the set of parking genetic operators, screening out parking feasible domain genetic operator calculation points using a target roulette algorithm.

[0095] It should be noted that the above explanations of the parking feasible region calculation method embodiment are also applicable to the parking feasible region calculation device of this embodiment, and will not be repeated here.

[0096] According to the parking feasible domain calculation device proposed in the embodiment of the present application, the constraints of the initial parking feasible domain are established based on the vehicle dynamics model, parking conditions and parking space restrictions, and the path planning function is established according to the vehicle dynamics model and the parking space position. The nonlinear optimization function is obtained by combining the constraints and the path planning function, and the nonlinear optimization function is solved based on the genetic algorithm to obtain the initial iteration point, and the path planning function is iteratively solved using the initial iteration point to calculate the final parking feasible domain. In this way, the problem of low calculation accuracy of the parking feasible domain during automatic parking is solved, the parking feasible domain is judged and optimized, and the accuracy and reliability of automatic parking are greatly improved.

[0097] Figure 8A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0098] A memory 801 , a processor 802 , and a computer program stored in the memory 801 and executable on the processor 802 .

[0099] When the processor 802 executes the program, the parking feasible region calculation method provided in the above embodiment is implemented.

[0100] Furthermore, the vehicle also includes:

[0101] The communication interface 803 is used for communication between the memory 801 and the processor 802 .

[0102] The memory 801 is used to store computer programs that can be executed on the processor 802 .

[0103] The memory 801 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0104] If the memory 801, the processor 802 and the communication interface 803 are implemented independently, the communication interface 803, the memory 801 and the processor 802 can be connected to each other through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0105] Optionally, in a specific implementation, if the memory 801, the processor 802 and the communication interface 803 are integrated on a chip, the memory 801, the processor 802 and the communication interface 803 can communicate with each other through an internal interface.

[0106] The processor 802 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0107] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which implements the above parking feasible domain calculation method when executed by a processor.

[0108] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0109] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0110] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0111] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.

[0112] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0113] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for calculating the feasible parking area. It is characterized in that The following steps are involved: Obtaining a vehicle dynamics model, parking conditions, parking space restrictions, parking space positions, and an initial parking feasible domain of the current vehicle; Establishing constraint conditions of the initial parking feasible domain according to the vehicle dynamics model, the parking conditions and the parking space restriction conditions, and establishing a path planning function according to the vehicle dynamics model and the parking space position; A nonlinear optimization function is obtained according to the constraint conditions and the path planning function, and the nonlinear optimization function is solved based on a preset genetic strategy to obtain an initial iteration point, and the path planning function is iteratively solved using the initial iteration point to calculate a final parking feasible domain.

2. The method according to claim 1, It is characterized in that After calculating the final parking feasible region, the method further includes: Based on the final parking feasible domain, the parking of the current vehicle is predicted, so as to complete the parking action of the current vehicle according to the prediction result.

3. The method according to claim 1, It is characterized in that The preset genetic strategy is: Obtain multiple initial target parking feasible regions; Generate multiple initial target parking feasible domain genetic operator populations, calculate the fitness of each initial target parking feasible domain genetic operator, and determine whether a preset iteration condition restriction is satisfied based on the fitness; If the preset iteration condition restriction is not satisfied, two parking feasible domains are selected based on a preset random roulette strategy to calculate genetic operator populations, and after optimizing the two parking feasible domains, it is re-determined whether the fitness satisfies the preset iteration condition restriction, until the fitness satisfies the preset iteration condition restriction, and the optimal parking feasible domain solution is obtained.

4. The method according to claim 3, It is characterized in that The optimizing process of calculating the genetic operator populations of the two parking feasible regions includes: Performing a copy operation on any one of the two parking feasible domains, and adding the copy result to the new population; Performing a mutation operation on the remaining parking feasible regions in the two parking feasible regions, and adding the mutation result to the new population; A crossover operation is performed on the two parking feasible regions, and the crossover result is added to the new population.

5. The method according to claim 3, It is characterized in that The preset random roulette strategy is: Determine a set of genetic operator points, wherein the set of points belongs to a set of genetic operator points that meet the parking feasible domain constraint condition; Based on the genetic operator set points and the preset random number, a set of parking genetic operators whose selection probability is greater than or equal to the preset random number is screened out from the set of genetic operator points that meet the parking feasible domain constraint condition, wherein the preset random number obeys a uniform distribution and is within an interval formed by zero and a function fitness probability calculated by the parking feasible domain genetic operator; Based on the set of parking genetic operators, the target roulette algorithm is used to screen out genetic operator calculation points of the parking feasible domain.

6. A parking feasible region calculation device, It is characterized in that include: An acquisition module, used to acquire a vehicle dynamics model, parking conditions, parking space restriction conditions, parking space position and an initial parking feasible domain of the current vehicle; A modeling module, configured to establish the constraint conditions of the initial parking feasible domain according to the vehicle dynamics model, the parking conditions and the parking space restriction conditions, and to establish a path planning function according to the vehicle dynamics model and the parking space position; A calculation module is used to obtain a nonlinear optimization function according to the constraint conditions and the path planning function, and solve the nonlinear optimization function based on a preset genetic strategy to obtain an initial iteration point, and use the initial iteration point to iteratively solve the path planning function to calculate a final parking feasible domain.

7. The device according to claim 6, It is characterized in that After calculating the final parking feasible region, the calculation module is further used to: Based on the final parking feasible domain, the parking of the current vehicle is predicted, so as to complete the parking action of the current vehicle according to the prediction result.

8. The device according to claim 6, It is characterized in that The preset genetic strategy is: Obtain multiple initial target parking feasible regions; Generate multiple initial target parking feasible domain genetic operator populations, calculate the fitness of each initial target parking feasible domain genetic operator, and determine whether a preset iteration condition restriction is satisfied based on the fitness; If the preset iteration condition restriction is not satisfied, two parking feasible domains are selected based on a preset random roulette strategy to calculate genetic operator populations, and after optimizing the two parking feasible domains, it is re-determined whether the fitness satisfies the preset iteration condition restriction, until the fitness satisfies the preset iteration condition restriction, and the optimal parking feasible domain solution is obtained.

9. A vehicle, It is characterized in that include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the parking feasible domain calculation method according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that The program is executed by a processor to implement the parking feasible region calculation method as described in any one of claims 1 to 5.