Automatic driving integrated planning control method

The integrated planning and control of genetic algorithms solves the structural complexity and error transmission problems of traditional methods, improves the driving accuracy and safety stability of autonomous driving vehicles, simplifies the calculation process, and avoids dependence on a large amount of training data.

CN118907138AInactive Publication Date: 2024-11-08CHANGJIANG INST OF TECH +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410797496.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-20
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional autonomous driving planning and control methods have complex structures, error conduction and frequency differences, which lead to poor quality of vehicle driving behavior. The existing integrated methods are complex in calculations and rely on a large amount of training data, affecting safety and stability.

Method used

Genetic algorithms are used for integrated planning control, and the initial population is generated by determining decision variables and individual structures. Based on the random inheritance algorithm, iterative evolution is used to obtain the optimal individual, calculate the vehicle's driving path point set and front wheel deflection control amount to avoid linearity simplify the calculation process.

Benefits of technology

It improves vehicle driving accuracy and safety stability, simplifies the planning control system structure, avoids error conduction and frequency differences, and reduces dependence on training data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118907138A_ABST
    Figure CN118907138A_ABST
Patent Text Reader

Abstract

The invention provides an automatic driving integrated planning control method, which comprises the steps of generating an initial population of a genetic algorithm, and encoding individuals of the initial population; decoding the current population, calculating a vehicle driving path point set represented by an individual, calculating an individual objective function value, selecting copy and crossover mutation operation, and obtaining an optimal individual through iterative evolution; and decoding the optimal individual into a front wheel deflection angle control increment sequence, calculating the front wheel deflection angle control quantity of the current control period according to the front wheel deflection angle control quantity of the previous control period, and outputting the front wheel deflection angle control quantity to the automatic driving vehicle for execution. According to the invention, an integrated planning control method can be provided for automatic driving, and the optimal front wheel deflection angle control quantity can be calculated by directly adopting an intelligent optimization method according to road environment information, a vehicle driving target and vehicle state information. Compared with a traditional method and an existing method, the method has the advantages that the planning control structure and the calculation process are simple, the vehicle driving precision is high, and the driving safety and stability are high.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving, and more specifically, to an integrated planning and control method for autonomous driving. Background Art

[0002] Path planning and tracking control is one of the key technologies for autonomous driving. It plans an obstacle avoidance path for the intelligent vehicle and controls the steering wheel angle or front wheel deflection angle to make the intelligent vehicle move along the path.

[0003] Most traditional autonomous driving planning and control methods adopt a "hierarchical" structure, that is, planning first and then controlling. In the planning layer, a smooth obstacle avoidance path is planned based on the road environment information and the vehicle's driving target; in the control layer, the planned path is used as the tracking target to calculate the front wheel deflection control amount. Finally, the front wheel deflection control amount is output to the vehicle to execute steering, so that the vehicle can drive safely according to the driving target. Patent papers on hierarchical planning and control include: CN201711081091.6, Paper: Li Jun, Li Guyue. Path planning and tracking control based on improved artificial potential field [J]. Journal of Chongqing Jiaotong University (Natural Science Edition), 2020, 39(09): 25-30+37, etc. Hierarchical planning and control has the following problems: (1) Since the autonomous driving planning and control system takes road environment information, vehicle status information and vehicle driving target as input, and calculates the vehicle steering wheel angle / front wheel deflection angle control amount output as the target. Therefore, the target path planned by the hierarchical path planning and control method is only an intermediate variable of the planning and control system, not a necessary link, and the structure of the hierarchical planning and control system is complex. (2) The error in the planning layer will be brought into the control layer, resulting in a deterioration in the quality of vehicle driving behavior. (3) The difference between the path planning frequency and the tracking control frequency is too large, resulting in insufficient stability in vehicle control.

[0004] In order to solve the problem of hierarchical planning and control, scholars have proposed an integrated planning and control method, that is, based on road environment information, vehicle status information and vehicle driving goals as input, the steering wheel angle or front wheel deflection angle of the intelligent vehicle is directly calculated through the integrated planning and control method, and then the front wheel deflection angle control amount is output to the vehicle for execution. One type of integrated planning and control method, such as patent CN202211649723.5, uses the vehicle dynamics equation based on linearized approximation to construct a constrained optimization problem and solve the optimal front wheel deflection angle control amount. The process is relatively complex to calculate, and the linearization and approximation of the vehicle dynamics equation will itself produce errors, so the control accuracy of the autonomous driving vehicle will be reduced. Another type of integrated planning and control method, such as patent: CN202311179912.5, uses a generative method and a brain-like neural circuit model to solve the problem of integrated planning and control of autonomous driving, but it is necessary to train the generative world model, which depends on a large amount of training data. If the amount of training data is insufficient, it will affect the safety and stability of autonomous driving. Summary of the invention

[0005] The present invention aims to solve the technical problems existing in the prior art and provides an integrated planning and control method for automatic driving, comprising:

[0006] Determine the decision variables of the genetic algorithm and the structure of the genetic algorithm individuals;

[0007] Generate an initial population consisting of P individuals, and encode each individual in the initial population based on the random inheritance algorithm;

[0008] Decode the current population based on the genetic algorithm, calculate the vehicle path point set represented by each individual in the current population, and calculate the individual objective function value. Based on the individual objective function value, select the individual for replication and crossover mutation operations, and obtain the optimal individual through iterative evolution;

[0009] Decode the optimal individual into a front wheel angle control increment sequence, calculate the front wheel angle control amount of the current control cycle based on the front wheel angle control amount of the previous control cycle, and output it to the autonomous driving vehicle for execution;

[0010] The decoding of the current population based on the genetic algorithm includes:

[0011] The decoding algorithm based on vehicle speed calculates the solution represented by each individual in the current population, and calculates the front wheel deflection angle control sequence represented by each individual in the current population;

[0012] The step of calculating the vehicle travel path point set represented by each individual in the current population includes:

[0013] According to the front wheel deflection angle control sequence represented by each individual, based on the nonlinear equation of vehicle kinematics or dynamics, a set of vehicle driving path points represented by each individual is calculated;

[0014] The calculating of the individual objective function value comprises:

[0015] Obtain road environment information and vehicle driving goals, and calculate the objective function value of each individual according to the front wheel deflection angle control sequence and vehicle driving path point set represented by each individual;

[0016] Based on the individual objective function value, the individual is selected for replication and crossover mutation operations, and the optimal individual is obtained through iterative evolution, including:

[0017] Determine whether the current population has reached the maximum number of iterations. If so, the iterative evolution ends and the optimal individual is output. If not, the fitness value of each individual is calculated based on the objective function value of each individual, and the individuals in the current population are selected and cross-mutated based on the fitness value of each individual, and then enter the next iterative evolution.

[0018] The present invention provides an integrated planning and control method for autonomous driving, which adopts an intelligent optimization method to solve the optimal control of the front wheel deflection control value, and the calculation process is simple; it adopts nonlinear vehicle kinematics / dynamics equations to improve the driving accuracy of the autonomous driving vehicle; it avoids dependence on a large amount of training data, and improves the safety and stability of vehicle driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of an integrated planning and control method for autonomous driving provided by the present invention;

[0020] Figure 2 The figure is a schematic diagram of the overall process of an integrated planning and control method for autonomous driving. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not subject to the constraints of the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0022] The problems of traditional hierarchical planning and control methods are as follows: (1) unnecessary intermediate variables are generated, and the structure is complex; (2) the errors in the planning layer are carried into the control layer, resulting in poor quality of vehicle driving behavior; (3) the difference between the path planning frequency and the tracking control frequency is too large, resulting in insufficient stability of vehicle control.

[0023] Problems with existing integrated planning and control methods: (1) Constructing and solving a constrained planning and control optimization problem requires a relatively complex calculation process; (2) Linearizing and approximating the vehicle kinematics / dynamics equations will itself produce errors, resulting in low control accuracy for autonomous driving vehicles; (3) The generative world model needs to be trained, which relies on a large amount of training data. If the training data is insufficient, the safety performance of autonomous driving will be affected.

[0024] Based on this, the present invention provides an integrated planning and control method for autonomous driving, see Figure 1 and Figure 2 , the control method mainly includes:

[0025] Step 1: Determine the decision variables of the genetic algorithm and the structure of the genetic algorithm individuals.

[0026] It is understandable that the present invention uses a genetic algorithm to solve the control sequence of the autonomous driving vehicle at the future moment. First, the decision variables of the genetic algorithm are respectively the front wheel deflection control increments of the autonomous driving vehicle at the first moment, the second moment, ..., and the Mth moment in the future, and the solution of the genetic algorithm is the front wheel deflection control increment sequence of the autonomous driving vehicle at the Mth moment in the future, and the sequence includes the front wheel deflection control increments from the first moment to the Mth moment in the future.

[0027] Then, the individual structure of the genetic algorithm is designed. Assume that the individual chromosome of the genetic algorithm is composed of M genes, and each gene is composed of N binary bits. The individual chromosome represents a solution of the genetic algorithm, and each gene of the individual represents a decision variable. A decision variable represents the front wheel angle control increment at a moment.

[0028] Step 2: Generate an initial population consisting of P individuals, and encode each individual of the initial population based on a random inheritance algorithm.

[0029] Specifically, a first subpopulation consisting of 2P / 3 individuals is generated based on a random algorithm, and the gene code of each individual in the first subpopulation is randomly assigned an initial value of 0 or 1.

[0030] A second subpopulation consisting of P / 3 individuals is generated based on the inheritance algorithm, and the gene code of each individual in the second subpopulation is assigned according to the gene code of the optimal individual in the previous control cycle.

[0031] The first subpopulation and the second subpopulation are combined to form an initial population consisting of P individuals. Specifically, the initial population can be represented as an array A of M rows and N columns, and each element in A is 0 or 1.

[0032] Step 3: decode the current population based on the genetic algorithm, calculate the vehicle travel path point set represented by each individual in the current population, and calculate the individual objective function value. Based on the individual objective function value, select replication and crossover mutation operations on the individual to obtain the optimal individual through iterative evolution.

[0033] Wherein, step 3 mainly includes the following steps:

[0034] Step 3.1: Use a decoding algorithm based on vehicle speed to calculate the solution represented by each individual in the population, and further calculate the front wheel deflection angle control sequence represented by each individual in the population.

[0035] It is understandable that step 3.1 includes the following steps:

[0036] (1) Extract the binary code corresponding to each individual decision variable. Taking the pth individual as an example, let the pth individual color code be Vp = a 11 a 12 a 13 …a 1N a 21 a 22 a 23 …a 2N ……a M1 a M2 a M3 …a MN , where a 11 、a 12、a 13 ,……,a MN are all 0 or 1, then the binary code of the mth decision variable is a m1 a m2 a m3 …a mN .

[0037] (2) Calculate the speed parameter based on the real-time speed of the autonomous driving vehicle. Specifically, the current speed of the autonomous driving vehicle is obtained as vm / s, and the speed parameter is When v <v min season When v>v max season When v min <v<v max season where v min is the minimum speed parameter, v max is the maximum speed parameter.

[0038] (3) The speed-based decoding algorithm is used to calculate the decision variables of each individual and the solution represented by each individual. Taking the pth individual as an example, the speed-based decoding algorithm is used to calculate each decision variable of the individual, and the mth decision variable Δδ in the pth individual is m For example, the calculation formula is:

[0039]

[0040] Where i represents the i-th bit of the binary code of the m-th decision variable, b m is the decimal number corresponding to the binary code of the mth decision variable, k v is the decision variable coefficient.

[0041] Solve each decision variable of the pth individual, and then obtain the solution represented by the pth individual: [Δδ1Δδ2Δδ3…Δδ m …Δδ M ].

[0042] (4) Further calculate the front wheel deflection control sequence represented by each individual. Taking the pth individual as an example, let the front wheel deflection control sequence represented by the pth individual be [δ1δ2δ3…δ m …δ M ], assuming that the front wheel deflection angle control value of the previous control cycle is δ0, then the front wheel deflection angle control value δ at the mth moment of the current control cycle is m The calculation formula is:

[0043]

[0044] Among them, j represents the jth moment, and when δ m >δ max When δ m =δ max , when δ m <-δ max When δ m =-δ max , where δ max Indicates the maximum limit of the front wheel deflection angle.

[0045] Step 3.2: According to the front wheel deflection angle control sequence represented by each individual, the vehicle driving path point set represented by each individual is calculated using the nonlinear equations of vehicle kinematics or dynamics.

[0046] Among them, step 3.2 mainly includes the following steps:

[0047] (1) According to the front wheel deflection control sequence represented by the individual and the real-time speed of the autonomous driving vehicle, the future control sequence represented by the individual is constructed. Taking the pth individual as an example, the front wheel deflection control sequence represented by the pth individual is [δ1δ2δ3…δ m …δ M ], the real-time speed of the autonomous driving vehicle is v. In the present invention, considering that the speed will not change much in the future, it is assumed that the speed from the first moment to the Mth moment in the future remains unchanged, which is v. According to the front wheel deflection angle control sequence represented by the pth individual and the real-time speed of the autonomous driving vehicle, the future control sequence represented by the pth individual is constructed as follows:

[0048]

[0049] (2) Based on the vehicle kinematics model or the vehicle dynamics model, a vehicle travel path point set prediction model is constructed. Specifically, the vehicle kinematics model or the vehicle dynamics model is described as a nonlinear equation:

[0050]

[0051] Where X(t)=[x(t),y(t),θ(t)] T , t is time, x(t), y(t), θ(t) are the horizontal coordinate, vertical coordinate and vehicle heading of the vehicle in the geodetic coordinate system, u(t) = [δ(t), v(t)] T , δ(t) is the front wheel deflection angle control value, and v(t) is the speed control value.

[0052] By discretizing the nonlinear equations of vehicle kinematics or dynamics, we can obtain:

[0053] X(k+1)=f(T,X(k),U(k));

[0054] Among them, T is the sampling period and k is the time.

[0055] According to the discretized nonlinear equation, the vehicle driving path point set prediction model is constructed as follows:

[0056]

[0057] (3) Based on the future control sequence represented by each individual and the vehicle driving path point set prediction model, predict the vehicle driving path point set represented by each individual. Taking the pth individual as an example, the future control sequence represented by the pth individual [U1 U2U3…U m …U M ] is brought into the vehicle driving path point set prediction model, and the horizontal and vertical coordinates and heading at the future moment are obtained by recursive method, and the vehicle driving path point set represented by the pth individual is:

[0058] Y=[X(2)X(3)X(4)…X(m)…X(M+1)];

[0059] Among them, X(m)=[x(m),y(m),θ(m)] T , are the horizontal coordinate, vertical coordinate and vehicle heading of the vehicle’s driving path point at time m respectively.

[0060] Step 3.3: Obtain road environment information and vehicle driving goals, and calculate the objective function value of each individual based on the front wheel deflection angle control sequence and vehicle driving path point set represented by each individual.

[0061] Among them, step 3.3 mainly includes the following steps:

[0062] (1) Obtain the current road environment information and vehicle driving goals of the autonomous driving vehicle, and use the artificial potential field method to calculate the first objective function value of each individual based on the vehicle driving path point set represented by the individual.

[0063] First, obtain the current road environment information of the autonomous driving vehicle, including lane line information, road boundary information, and obstacle information, and obtain the vehicle driving target as the pre-set global path Y r .

[0064] Then, the artificial potential field method is used to calculate the first objective function value of each individual. Taking the pth individual as an example, the first objective function value of the pth individual is calculated, which specifically includes the following process:

[0065] The lane potential field strength value of the vehicle path point set represented by the pth individual is calculated as:

[0066]

[0067] Among them, A lane represents the central potential field coefficient, σ lane is the lane width parameter, d 1m Represents the distance between the mth path point in the vehicle travel path point set Y represented by the pth individual and the nearest lane line.

[0068] The road boundary potential field strength value of the vehicle path point set represented by the pth individual is calculated as:

[0069]

[0070] Among them, A border represents the boundary potential field coefficient, d 2m Represents the distance from the mth path point in the vehicle travel path point set Y represented by the pth individual to the nearest boundary line.

[0071] The potential field strength value of the target point of the vehicle path point set represented by the pth individual is calculated as:

[0072] U goal =ε-kd3;

[0073] Among them, ε and k are the potential field parameters of the target point, and the values ​​of ε and k ensure that U goal >0, d3 represents the Mth path point of the vehicle driving path point set Y represented by the pth individual and the vehicle's current position in the vehicle driving target Y r The distance between the projection points on .

[0074] The obstacle potential field strength value of the vehicle path point set represented by the pth individual is calculated as:

[0075]

[0076] Among them, d0 is the influence range of the obstacle potential field, and d4 represents the minimum distance from the vehicle path point set Y represented by the pth individual to the obstacle.

[0077] Calculate the first objective function value of the pth individual:

[0078] J1=U lane +U road +U goal +U obs .

[0079] (2) Obtain the vehicle driving target and calculate the second objective function value of each individual.

[0080] First, the vehicle driving target is obtained as a pre-set global path Y r .

[0081] Then, the second objective function value of each individual is calculated. Taking the p-th individual as an example, the second objective function value of the p-th individual is calculated, which specifically includes the following process:

[0082] Calculate the projection point set Y = [X(2)X(3)X(4)…X(m)…X(M+1)] of the driving path represented by the pth individual on the vehicle driving target as Y r =[X r (2)X r (3)X r (4)…X r (m)…X r (M+1)], where X r (m) = [x r (m),y r (m),θ r (m)] T .

[0083] Calculate the second objective function value of the pth individual:

[0084]

[0085] Among them, α m , β m They are the position error weight coefficient and heading error weight coefficient at time m respectively.

[0086] (3) According to the front wheel deflection control sequence represented by the individual, the third objective function value of each individual is calculated. Specifically, taking the pth individual as an example, the front wheel deflection control sequence represented by the pth individual is [δ1δ2δ3…δ m …δ M ], calculate the third objective function value of the pth individual:

[0087]

[0088] Among them, q m Controls the incremental weight coefficient at the mth moment.

[0089] (4) Calculate the objective function value of each individual according to the first objective function value, the second objective function value and the third objective function value of each individual. Specifically, taking the pth individual as an example, the first objective function value, the second objective function value and the third objective function value of the pth individual are J1, J2 and J3 respectively, and the objective function value of the pth individual is calculated as:

[0090] J = η1J1 + η2J2 + η3J3;

[0091] Among them, η1, η2, and η3 are the weight coefficients of the first objective function value, the second objective function value, and the third objective function respectively.

[0092] Step 3.4: Determine whether the current population has reached the maximum number of iterations. If so, the iterative evolution ends and the optimal individual is output. If not, the fitness value of each individual is calculated, selection and crossover operations are performed, and the next iterative evolution begins.

[0093] (1) Determine whether the number of iterative evolutions of the current population has reached the set maximum number of iterations. If so, go to (4); if not, go to (2).

[0094] (2) According to the objective function value of all individuals in the current population, the fitness value of each individual is calculated. Taking the pth individual as an example, the fitness function value calculation formula of the pth individual is:

[0095]

[0096] (3) Using the roulette algorithm, select P individuals from the current population, and use the crossover operator and mutation operator of the genetic algorithm to perform crossover and mutation operations on the selected P individuals to obtain the next generation population, and return to step 3.1 for the next iterative evolution.

[0097] (4) From the current population, select the individual with the smallest objective function value (i.e., the largest fitness value) as the optimal individual, and output the binary code of the optimal individual.

[0098] Step 4: decode the optimal individual into a front wheel angle control increment sequence, calculate the front wheel angle control amount of the current control cycle based on the front wheel angle control amount of the previous control cycle, and output it to the autonomous driving vehicle for execution.

[0099] Wherein, step 4 mainly includes the following steps:

[0100] (1) The optimal individual is decoded into the front wheel deflection angle control increment sequence [Δδ1 * Δδ2 * Δδ3 * …Δδ m * …Δδ M * ].

[0101] (2) Obtain the front wheel deflection control amount δ0 of the previous control cycle, and calculate the control increment Δδ1 according to the first front wheel deflection angle * , calculate the first front wheel deflection control amount of the current control cycle as: δ * =δ0+Δδ1 * As the front wheel deflection control amount of the current control cycle, and when δ * >δ max When δ *=δ max , when δ * <-δ max When δ * =-δ max .

[0102] (3) * Output to the autonomous vehicle for execution.

[0103] Compared with the traditional hierarchical planning and control method, the present invention has the following benefits: (1) no intermediate variables are generated, and the planning control system structure is simple; (2) the integrated planning and control structure will not cause the error of the planning layer to be transmitted to the control layer, and the vehicle driving accuracy is improved; (3) there is no frequency difference between planning and control, and the vehicle control is stable.

[0104] Compared with the existing integrated planning and control methods, the present invention has the following benefits: (1) It adopts an intelligent optimization method to solve the optimal control of the front wheel deflection control value, and the calculation process is simple; (2) It adopts nonlinear vehicle kinematics / dynamics equations without linear approximation processing, and the driving accuracy of the autonomous driving vehicle is improved; (3) It avoids dependence on a large amount of training data, thereby improving the safety and stability of vehicle driving.

[0105] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0106] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0110] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0111] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An integrated planning and control method for automatic driving, characterized in that: include: Determine the decision variables of the genetic algorithm and the structure of the genetic algorithm individuals; Generate an initial population consisting of P individuals, and encode each individual in the initial population based on the random inheritance algorithm; Decode the current population based on the genetic algorithm, calculate the vehicle path point set represented by each individual in the current population, and calculate the individual objective function value. Based on the individual objective function value, select the individual for replication and crossover mutation operations, and obtain the optimal individual through iterative evolution; Decode the optimal individual into a front wheel angle control increment sequence, calculate the front wheel angle control amount of the current control cycle based on the front wheel angle control amount of the previous control cycle, and output it to the autonomous driving vehicle for execution; The decoding of the current population based on the genetic algorithm includes: The decoding algorithm based on vehicle speed calculates the solution represented by each individual in the current population, and calculates the front wheel deflection angle control sequence represented by each individual in the current population; The step of calculating the vehicle travel path point set represented by each individual in the current population includes: According to the front wheel deflection angle control sequence represented by each individual, based on the nonlinear equation of vehicle kinematics or dynamics, a set of vehicle driving path points represented by each individual is calculated; The calculating of the individual objective function value comprises: Obtain road environment information and vehicle driving goals, and calculate the objective function value of each individual according to the front wheel deflection angle control sequence and vehicle driving path point set represented by each individual; Based on the individual objective function value, the individual is selected for replication and crossover mutation operations, and the optimal individual is obtained through iterative evolution, including: Determine whether the current population has reached the maximum number of iterations. If so, the iterative evolution ends and the optimal individual is output. If not, the fitness value of each individual is calculated based on the objective function value of each individual, and the individuals in the current population are selected and cross-mutated based on the fitness value of each individual, and then enter the next iterative evolution.

2. The integrated planning and control method for autonomous driving according to claim 1, characterized in that: The decision variables of the genetic algorithm are the front wheel deflection angle control increments of the autonomous driving vehicle at the first moment, the second moment, ..., and the Mth moment in the future, and the solution of the genetic algorithm is the front wheel deflection angle control increment sequence of the autonomous driving vehicle at the Mth moment in the future; Each individual chromosome in the genetic algorithm is composed of M genes, each gene is composed of N binary bits, each individual chromosome represents a solution of the genetic algorithm, and each gene of the individual represents a decision variable, that is, the front wheel deflection angle control increment at a certain moment from the first moment to the Mth moment in the future.

3. The integrated planning and control method for autonomous driving according to claim 2, characterized in that: The vehicle speed-based decoding algorithm calculates the solution represented by each individual in the current population, and calculates the front wheel deflection angle control sequence represented by each individual in the current population, including: Extract the binary code corresponding to each individual decision variable; Calculate vehicle speed parameters according to the real-time vehicle speed of the autonomous driving vehicle; The decoding algorithm based on vehicle speed calculates the decision variables of each individual and the solution represented by each individual, that is, the front wheel deflection angle control increment sequence represented by each individual; According to the front wheel deflection angle control amount of the previous control cycle and the solved front wheel deflection angle control increment sequence, the front wheel deflection angle control sequence represented by each individual in the current population is calculated.

4. The integrated planning and control method for autonomous driving according to claim 3, characterized in that: The step of extracting the binary code corresponding to each individual decision variable includes: Suppose the coloring code of the pth individual is Vp=a 11 a 12 a 13 …a 1N a 21 a 22 a 23 …a 2N ……a M1 a M2 a M3 …a MN , where a 11 、a 12 、a 13 ,……,a MN are all 0 or 1, then the binary code of the mth decision variable of the pth individual is a m1 a m2 a m3 …a mN ; The speed-based decoding algorithm calculates the decision variables of each individual and the solution represented by each individual, including: The mth decision variable Δδ of the pth individual m The calculation formula is: Where i represents the i-th bit of the binary code of the m-th decision variable, b m is the decimal number corresponding to the binary code of the mth decision variable, k v is the decision variable coefficient, is the vehicle speed parameter, where: When the current speed of the autonomous driving vehicle v <v min season When v>v max season When v min <v<v max season v min is the minimum speed parameter, v max is the maximum speed parameter; Get the front wheel deflection control increment sequence [Δδ1Δδ2Δδ3…Δδ m …Δδ M ]; The method of calculating the front wheel deflection angle control sequence represented by each individual in the current population according to the front wheel deflection angle control amount of the previous control cycle and the solved front wheel deflection angle control increment sequence comprises: According to the front wheel deflection angle control increment sequence [Δδ1Δδ2Δδ3…Δδ m …Δδ M ] and the front wheel deflection control value δ0 of the previous control cycle, calculate the mth front wheel deflection control value δ m : Among them, when δ m >δ max When δ m =δ max , when δ m <-δ max When δ m =-δ max , δ max Indicates the maximum limit of the front wheel deflection angle; The front wheel deflection control sequence represented by the pth individual is obtained as [δ1δ2δ3…δ m …δ M ].

5. The integrated planning and control method for autonomous driving according to claim 1, characterized in that: The method of calculating the vehicle driving path point set represented by each individual according to the front wheel deflection angle control sequence represented by each individual and based on the nonlinear equation of vehicle kinematics or dynamics includes: According to the front wheel deflection angle control sequence represented by the individual and the real-time speed of the autonomous driving vehicle, the future control sequence represented by the individual is constructed; According to the real-time position, heading and future control sequence represented by the autonomous driving vehicle, a vehicle travel path point set prediction model is constructed based on the vehicle kinematic model or vehicle dynamics model; According to the future control sequence represented by each individual and the vehicle driving path point set prediction model, the vehicle driving path point set represented by each individual is calculated, and the vehicle driving path point set includes the predicted position information and heading information of each driving path point.

6. The integrated planning and control method for autonomous driving according to claim 5, characterized in that: The future control sequence represented by the individual is: Among them, [δ1δ2δ3…δ m …δ M ] represents the front wheel deflection angle control sequence represented by the individual, and v is the real-time speed of the autonomous driving vehicle; The method constructs a vehicle travel path point set prediction model based on the vehicle kinematics model or vehicle dynamics model according to the real-time position and heading of the autonomous driving vehicle and the future control sequence represented by the individual, including: The vehicle kinematics model or vehicle dynamics model is described as a nonlinear equation: Among them, X(t)=[x(t),y(t),θ(t)] T , t is time, x(t), y(t), θ(t) are the horizontal coordinate, vertical coordinate and vehicle heading of the autonomous driving vehicle in the geodetic coordinate system, u(t) = [δ(t), v(t)] T , δ(t) is the front wheel deflection angle control value, v(t) is the speed control value; Discretizing the nonlinear equations of vehicle kinematics or vehicle dynamics yields: X(k+1)=f(T,X(k),U(k)); Among them, T is the sampling period, k is the time; According to the discretized nonlinear equation, the vehicle driving path point set prediction model is constructed as follows: The method of calculating the vehicle travel path point set represented by each individual according to the future control sequence represented by each individual and the vehicle travel path point set prediction model includes: The future control sequence corresponding to the individual [U1 U2U3…U m …U M ] is brought into the vehicle driving path point set prediction model, and the horizontal and vertical coordinates and heading of the autonomous driving vehicle in the future M moments are obtained through the recursive method. The vehicle driving path point set represented by the pth individual is: Y=[X(2)X(3)X(4)…X(m)…X(M+1)] Among them, X(m)=[x(m),y(m),θ(m)] T , which are the horizontal coordinate, vertical coordinate and vehicle heading of the vehicle’s driving path point at time m respectively.

7. The integrated planning and control method for autonomous driving according to claim 1, characterized in that: The road environment information and the vehicle driving target are obtained, and the objective function value of each individual is calculated according to the front wheel deflection angle control sequence represented by each individual and the vehicle driving path point set, including: The road environment information and vehicle driving target of the autonomous driving vehicle are obtained respectively. The road environment information includes lane line information, road boundary information and obstacle information. The vehicle driving target is a pre-set global path Y. r ; According to the lane line information, road boundary information and obstacle information, the first objective function value of each individual is calculated based on the artificial potential field method; Calculate the second objective function value of each individual according to the vehicle driving path point set and the vehicle driving goal represented by each individual; Calculate the third objective function value of each individual according to the front wheel deflection angle control sequence represented by the individual; Based on the first objective function value, the second objective function value and the third objective function value of each individual, the objective function value of each individual is calculated.

8. The integrated planning and control method for autonomous driving according to claim 7, characterized in that: The first objective function value of each individual is calculated based on the artificial potential field method according to the lane line information, the road boundary information and the obstacle information, including: The lane potential field strength value of the vehicle path point set represented by the pth individual is calculated as: Among them, A lane represents the central potential field coefficient, σ lane is the lane width parameter, d 1m Represents the distance between the mth path point of the vehicle travel path point set Y represented by the pth individual and the nearest lane line; The road boundary potential field strength value of the vehicle path point set represented by the pth individual is calculated as: Among them, A border represents the boundary potential field coefficient, d 2m Indicates the distance from the mth path point of the vehicle travel path point set Y represented by the pth individual to the nearest boundary line; The potential field strength value of the target point of the vehicle path point set represented by the pth individual is calculated as: U goal =e-kd3; Among them, ε and k are the potential field parameters of the target point, and the values ​​of ε and k ensure that U goal >0, d3 represents the Mth path point of the vehicle driving path point set Y represented by the pth individual and the vehicle's current position in the vehicle driving target Y r The distance between the projection points on ; The obstacle potential field strength value of the vehicle path point set represented by the pth individual is calculated as: Among them, d0 is the influence range of the obstacle potential field, d4 ​​represents the minimum distance from the vehicle path point set Y represented by the pth individual to the obstacle; Calculate the first objective function value of the pth individual: J1=U lane +U road +U goal +U obs 。 9. The integrated planning and control method for autonomous driving according to claim 7, characterized in that: The calculating of the second objective function value of each individual according to the vehicle driving path point set represented by each individual and the vehicle driving goal includes: Calculate the projection point set Y = [X(2)X(3)X(4)…X(m)…X(M+1)] of the driving path represented by the pth individual on the vehicle driving target as Y r =[X r (2)X r (3)X r (4)…X r (m)…X r (M+1)], where X r (m) = [x r (m),y r (m),θ r (m)] T ; Calculate the second objective function value of the pth individual: Among them, α m , β m are the position error weight coefficient and heading error weight coefficient at time m respectively; The calculating the third objective function value of each individual according to the front wheel deflection angle control sequence represented by the individual includes: Among them, J3 is the third objective function value, q m is the control increment weight coefficient at the mth moment, [δ1δ2δ3…δ m …δ M ] represents the front wheel angle control sequence represented by the individual.

10. The integrated planning and control method for autonomous driving according to claim 7, characterized in that: The step of calculating the objective function value of each individual based on the first objective function value, the second objective function value and the third objective function value of each individual comprises: J = η1J1 + η2J2 + η3J3; Among them, J1, J2, and J3 are the first objective function value, the second objective function value, and the third objective function value of the pth individual, respectively, and η1, η2, and η3 are the weight coefficients of the first objective function value, the second objective function value, and the third objective function, respectively.