Path tracking control method for four-wheel independent steering and driving vehicle based on composite control framework

Abstract

The application relates to a four-wheel independent steering and driving vehicle path tracking control method based on a composite control framework, which comprises the following steps: step 1, a seven-degree-of-freedom dynamics model of the four-wheel independent steering and driving vehicle is established, and high-performance path tracking is realized by reasonably allocating input parameters; step 2, a deep reinforcement learning transition model is established, the input parameters of the vehicle state and an auxiliary controller are defined, and a system reward function is designed to optimize the path tracking control performance; step 3, a twin-delay deep deterministic policy gradient algorithm is selected to realize deep reinforcement learning, the overestimation problem of the algorithm is reduced, and the stability of learning is improved; step 4, a double-layer control architecture is adopted, the upper-layer controller calculates generalized control variables based on nonlinear model predictive control, and the lower-layer controller distributes the generalized control variables to each wheel through sequential quadratic programming to solve the lateral and longitudinal coupling problem of the vehicle in path tracking control; and step 5, the auxiliary controller based on deep reinforcement learning compensates the control variables according to external disturbances and vehicle states in real time, so that the control precision and stability of path tracking are improved. Through the above method, stable control of the four-wheel independent steering and driving vehicle path tracking in a complex environment is realized, and the adaptability of the system to external disturbances is enhanced.

Description

Technical Field

[0001] The present invention relates to a path tracking control method for a four-wheel independent steering and drive vehicle. The specific design utilizes an auxiliary controller based on deep reinforcement learning to implement a compensation control method for control variables. Background Art

[0002] Four-wheel independent steering (FWS) vehicles, with independently controlled steering angles and torques at each wheel, offer exceptionally high path-tracking flexibility, significantly enhancing their adaptability and robustness in complex environments. However, the complex nonlinear characteristics of FWS vehicles pose numerous challenges in designing path-tracking controllers with fast response, high tracking accuracy, and strong anti-interference capabilities. In recent years, control strategies and algorithms have been developed for the path-tracking control of FWS vehicles. Key control strategies include: first, lateral and longitudinal decoupling control, enabling tracking of different targets through independent control circuits; and second, a hierarchical control structure, where an upper-level controller calculates the generalized forces required by the vehicle, and a lower-level controller precisely distributes these forces to each wheel. Commonly used control algorithms include sliding mode control, model predictive control, proportional-integral-derivative control, and their improved methods. Traditional path-tracking control methods still have limitations when addressing the complex nonlinear dynamics and lateral-longitudinal coupling characteristics of FWS vehicles. First, model-based control methods are difficult to cope with external disturbances and drastic environmental changes; second, numerical errors may gradually amplify in recursive calculations. In the case of external disturbances, the error accumulation problem will make it difficult to ensure the stability of path tracking control. Summary of the Invention

[0003] The present invention aims to overcome the above-mentioned shortcomings of the prior art and proposes a path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework.

[0004] The present invention's path tracking control method for a four-wheel independent steering and drive vehicle based on a composite control framework combines a model-based two-layer controller with a model-free deep reinforcement learning-assisted controller. The deep reinforcement learning-assisted controller compensates for the control variables of the two-layer controller to achieve accurate tracking of the control target. Specific implementation steps include:

[0005] Step 1: Establish a dynamic model of a four-wheel independent steering and drive vehicle;

[0006] Step 2: Establish a deep reinforcement learning transition model; the state vector s of the four-wheel independent steering and driving vehicle includes the measurable state o and the random disturbance d;

[0007] Step 3: Select the twin delayed deep deterministic policy gradient algorithm to implement deep reinforcement learning;

[0008] Step 4, a double-layer control architecture is used to solve the lateral and longitudinal coupling problem of four-wheel independent steering and driving vehicles in path tracking control, which is composed of an upper controller based on nonlinear model predictive control and a lower controller based on sequential quadratic programming, the upper controller calculates the current state error e of the vehicle through the ideal path tracking state information and the current state of the vehicle, and solves the optimal control increment through the nonlinear model predictive controller; the lower controller solves the cost function through the sequential quadratic programming, and distributes the generalized control variable u of the upper controller u c + to calculate the lateral force and longitudinal force of each wheel at the next moment;

[0009] Step 5, a composite control framework is used to control the vehicle through two control loops.

[0010] The proposed composite control framework controls the vehicle through two control loops: a double-layer control loop and a model-free deep reinforcement learning auxiliary control loop. The vehicle state information is transmitted to the upper controller to calculate the required generalized force at the next moment. At the same time, a deep reinforcement learning neural network is used as an auxiliary controller to generate an auxiliary control term according to the vehicle state information and external disturbances to compensate for the upper layer control variable. The lower controller is responsible for distributing the generalized force transmitted by the upper controller, and finally gives the control variable under the composite control framework.

[0011] Compared with the prior art, the advantages of the present application are: a composite control architecture based on a double-layer controller and a deep reinforcement learning auxiliary controller is proposed, the double-layer controller controls the closed loop of the four-wheel independent steering and driving vehicle, and the deep reinforcement learning auxiliary controller adjusts the wheel force according to the external disturbance and the vehicle state. The composite control framework improves the adaptability of the control system to external environmental changes and improves the control accuracy and stability of path tracking on the basis of stable control of four-wheel independent steering and driving vehicle path tracking. BRIEF DESCRIPTION OF DRAWINGS

[0012] Fig. 1(a) and Fig. 1(b) are schematic diagrams of the dynamics model of the four-wheel independent steering and driving vehicle of the present application: Fig. 1(a) is a vehicle body dynamics model, and Fig. 1(b) is a tire dynamics model;

[0013] Figure 2 is a schematic diagram of the composite control framework of the present application;

[0014] Figure 3 is a path tracking average absolute error (M AE) comparison diagram of the composite controller (TDC) and the double-layer controller (MLC) based on the twin delay deep deterministic policy gradient algorithm of the present application;

[0015] Figure 4This is a comparison chart of the path tracking maximum error index (MAX) of the composite controller (TDC) and the dual-layer controller (MLC) based on the twin delayed deep deterministic policy gradient algorithm;

[0016] Figure 5 This is a comparison chart of the path tracking performance of the composite controller (TDC) and the dual-layer controller (MLC) based on the twin delayed deep deterministic policy gradient algorithm and the reference path. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is further described below with reference to the accompanying drawings.

[0018] The present invention is based on a composite control architecture based on a two-layer controller and a deep reinforcement learning-assisted controller. This architecture uses a two-layer controller for closed-loop control of the four-wheel independent steering and drive vehicle, and uses a deep reinforcement learning controller to assist in adjusting the wheel forces based on external disturbances and vehicle status. The composite control strategy improves the control system's adaptability to changes in the external environment while ensuring the stability of the four-wheel independent steering and drive vehicle path tracking controller, and further improves the path tracking control accuracy and stability under complex external interference conditions. The specific design steps are:

[0019] The present invention's path tracking control method for a four-wheel independent steering and drive vehicle based on a composite control framework combines a model-based two-layer controller with a model-free deep reinforcement learning-assisted controller. The deep reinforcement learning-assisted controller compensates for the control variables of the two-layer controller, achieving precise tracking of the control target. The specific implementation steps include:

[0020] Step 1: Establish the dynamic model of the four-wheel independent steering and driving vehicle. As shown in Figure 1(a), OXYZ and O v X v Y v Z v The four tires are indexed as i = fl, fr, rl, rr respectively. In the tire coordinate system, the longitudinal force, lateral force, vertical force and steering angle of each tire are respectively represented by F. l,i ,F L,i ,F N,i and δ i The dynamic equation of the vehicle body can be written as:

[0021]

[0022] Where M and I γ represents the vehicle mass and moment of inertia, (Lbi ,l bi ) represents the distance between the tire and the vehicle's center of gravity, and They represent the vehicle's longitudinal acceleration, lateral acceleration, and yaw angular acceleration, respectively.

[0023] In the design process of the path tracking controller, the longitudinal force and lateral force of each tire are determined by adjusting the steering angle and wheel torque, as shown in Figure 1(b). In the tire coordinate system, the present invention uses the magic formula to represent the nonlinear characteristics of the tire model.

[0024] y=Dsin(Carctan{Hx-E[Hx-arctan(Hx)]}) (2)

[0025] Where y represents the longitudinal force or lateral force, x represents the longitudinal slip rate or slip angle, and the peak coefficient D, stiffness coefficient H, shape coefficient C, and curvature coefficient E are the fitting coefficients of the empirical formula.

[0026] Longitudinal slip rate λ i The calculation formula is

[0027]

[0028] Where R w and ω i represents the tire radius and rotational angular velocity, u w,i The actual speed at the center of the tire. The tire speed can be calculated using the following formula

[0029]

[0030] The steering angle and torque of the tire can be expressed as

[0031]

[0032] Where, T i is the driving torque, f w is the rolling friction coefficient, I w is the moment of inertia of the tire.

[0033] Based on the above analysis, we established a vehicle dynamics model with seven degrees of freedom. By reasonably allocating input parameters, we hope to effectively control the vehicle state and achieve high-performance path tracking. The control input matrix of the vehicle dynamics model is

[0034] U=[T fl T fr T rl T rr δ fl δ fr δrl δ rr ] (6)

[0035] Wherein, the matrix U includes the driving torque and steering angle of the left front wheel, right front wheel, left rear wheel and right rear wheel of the vehicle. The vehicle state matrix is ​​defined as

[0036] X=[v x v y γ] (7)

[0037] In order to facilitate the design of auxiliary controllers based on deep reinforcement learning, the dynamic equations of the four-wheel independent steering and driving vehicle can be written in nonlinear matrix form

[0038]

[0039] Where U represents the control output matrix of the two-layer controller, Represents the vehicle acceleration disturbance matrix, C represents the mapping matrix between the upper controller and the lower controller control quantity, and B represents the vehicle mass matrix. In order to truly reflect the impact of external disturbances on vehicle dynamics, we introduce the control-end disturbance matrix, which represents the lateral and longitudinal disturbance forces acting on each wheel.

[0040]

[0041] The external disturbance in equation (9) can be transferred to the vehicle body dynamics model through the tire dynamics model to determine the vehicle body acceleration disturbance matrix:

[0042] Step 2: Establish a deep reinforcement learning transition model. The state vector s of a four-wheeled independently steered and driven vehicle consists of a measurable state o and a random disturbance d. This patent uses deep reinforcement learning as an auxiliary controller rather than directly controlling the vehicle end-to-end, assuming that the disturbance d is known. During deep reinforcement learning training, the motion control process of the four-wheeled independently steered and driven vehicle is written as: Since the auxiliary controller based on deep reinforcement learning needs to control the vehicle in real time according to the vehicle status, the following optimization problem can be constructed:

[0043]

[0044] Where a=[F ax F ay M aφ ] is the auxiliary controller action based on deep reinforcement learning, which represents the auxiliary control input of the driving torque and steering angle of each wheel, f H (·) represents the four-wheel independent steering and driving vehicle dynamics function, f a (·) represents the auxiliary controller, fuc (·) represents the upper controller, f l (·) represents the lower-level controller, X * represents the reference path state of the longitudinal velocity, lateral velocity, and yaw velocity obtained from the lateral displacement y and yaw angle γ by differential operation.

[0045] The state of the deep reinforcement learning controller is designed to be in Indicates the error between the current actual vehicle state and the ideal state, represents the longitudinal velocity error, e Y Represents the lateral displacement error, e φ Indicates the heading angle error, u c =[F cx F cy M cφ ] represents the generalized force of the control end.

[0046] The reward function of the system is designed as:

[0047]

[0048] Where K r Represents the positive gain parameter of the path tracking error, K f Positive gain parameter representing continuous control smoothness, e ub Indicates the limit value of tracking error, K b It represents the penalty factor. It indicates the changing speed of the auxiliary control variable, and its purpose is to ensure that the control process has a certain smoothness.

[0049] Step 3: Select the Twin Delayed Deep Deterministic Policy Gradient algorithm to implement deep reinforcement learning. Twin Delayed Deep Deterministic Policy Gradient (TD3) is a reinforcement learning algorithm based on the Deep Deterministic Policy Gradient (DDPG) algorithm, which is used to solve the problem of training instability in continuous action spaces. TD3 reduces the over-estimation problem in traditional algorithms and improves learning stability and performance by introducing mechanisms such as dual Q networks, delayed policy updates, and target action noise. Its core idea is to use the experience replay pool to store interaction samples with the environment and break the temporal correlation through random sampling, so as to train the policy and Q network more efficiently.

[0050] Before formal training begins, the agent interacts with the environment by randomly selecting actions to accumulate sufficient initial experience. This step ensures that there are enough diverse samples in the experience replay pool for subsequent training. For the initial time step t, the agent is in each state s. t Next, randomly select action a t , record the immediate reward r after executing the action t and update status s t+1 , and the new transfer sample (s t ,a t ,r t ,s t+1 ) is stored in the experience replay pool Ω until enough samples are stored in the experience pool. Initialize the policy network π θ , two Q networks and and its corresponding target network π θ '、 and The initial parameters of the target network are the same as those of the current network. The experience replay pool Ω has already started collecting samples in the first step, and these samples will be used in subsequent training stages.

[0051] Randomly sample a batch of samples (s t ,a t ,r t ,s t+1 ). The experience replay pool breaks the temporal correlation of samples by storage and random sampling, effectively reducing the problem of policy overfitting. Then, using the target policy network π θ 'Generate the next state s t+1 The target action a t+1 , and introduce noise N t Smoothing to prevent the policy network from overfitting the action

[0052] a t+1 =π θ ′(s t+1 )+N t (12)

[0053] Using Target Q Network and Calculate the target Q value and take its minimum value to reduce the overestimation problem of the Q value

[0054]

[0055] Where y t is the target Q value, r t For immediate rewards.

[0056] The loss function is constructed as the mean squared error between the Q-value estimate and the target Q-value: Stochastic gradient descent is used to minimize the loss function of the two Q networks. Through the optimization of the loss function, TD3 will gradually adjust the parameters φ1 and φ2 of the Q network to improve the accuracy of the Q value estimation. TD3 introduces a delayed policy update mechanism, that is, the policy network π is updated once after each two Q networks are updated. θ The goal of the policy network is to find the action that maximizes the Q value

[0057] (14) In the formula, π θ TD3 uses delayed updates to optimize the policy network based on a more stable Q-value estimation, avoiding premature reliance on inaccurate Q-values ​​by the policy network. It uses a soft update method to update the parameters of the target Q network and the target policy network, so that the two target networks are updated smoothly from the main network.

[0058]

[0059] Where τ is the soft update step size. After training is completed, the policy network π θ The optimal strategy has been optimized. At this point, for a given state, the optimal action a = π can be directly calculated through the strategy network. θ (s), the action a is used to guide the agent’s actual decision-making in the environment.

[0060] Step 4: To solve the lateral and longitudinal coupling problem of the four-wheel independent steering and drive vehicle in path tracking control, the patent of this invention adopts a two-layer control architecture, which consists of an upper controller based on nonlinear model predictive control and a lower controller based on sequential quadratic programming. The upper controller calculates the current state error e of the vehicle through the ideal path tracking state information and the current state of the vehicle, and solves the optimal control increment through the nonlinear model predictive controller.

[0061]

[0062] In the formula, e(t k+n |t k ) represents time t k The state error, Q (k) and R (k) Denote the error weight matrix and the control increment weight matrix respectively, and ε denotes the relaxation factor to avoid the inability to find a feasible solution. The generalized control quantity is calculated by the cost function to calculate the state of the vehicle at the next moment

[0063] u c =[F cx F cy M cφ] (17)

[0064] Where, F cx 、F cy and M cφ They represent the generalized longitudinal force, lateral force and yaw moment of the vehicle body respectively.

[0065] To balance and minimize wheel utilization, the lower controller solves the cost function through sequential quadratic programming and allocates the generalized control quantity u=u of the upper controller. c +a, to calculate the lateral and longitudinal forces of each wheel at the next moment

[0066]

[0067] Where w1 and w2 represent the tire utilization minimization weight coefficient and tire utilization balance weight coefficient, respectively. The tire dynamics model calculates the lateral and longitudinal forces on each wheel as angle and torque, respectively. This determines the vehicle's terminal control variables at the next moment and transmits them to the vehicle body for closed-loop path tracking control.

[0068] In step 5, the composite control framework proposed in this patent controls the vehicle through two control loops: a two-layer control loop and a model-free deep reinforcement learning auxiliary control loop. The vehicle state information is transmitted to the upper-level controller to calculate the generalized force required at the next moment. At the same time, a deep reinforcement learning neural network is used as an auxiliary controller to generate auxiliary control items based on the vehicle state information and external disturbances to compensate for the upper-level control variables. The lower-level controller is responsible for distributing the generalized forces transmitted by the upper-level controller and finally providing the control variables under the composite control framework.

[0069] The content of the implementation examples in this specification is merely a description of the control framework and control process. The scope of protection of this invention should not be considered limited to the specific forms and parameters described in the implementation examples. In addition, the scope of protection of this invention also includes branching algorithms for different deep reinforcement learning.

Claims

1. A path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework, characterized in that: The model-based two-layer controller is combined with a model-free deep reinforcement learning-assisted controller. The deep reinforcement learning-assisted controller is used to compensate the control variables of the two-layer controller to achieve accurate tracking of the control target. Specific implementation steps include: Step 1: Establish a dynamic model of a four-wheel independent steering and drive vehicle; Step 2: Establish a deep reinforcement learning transition model; the state vector of the four-wheel independent steering and driving vehicle Including measurable status and random interference Specifically, during the deep reinforcement learning training process, the motion control process of the four-wheel independent steering and driving vehicle is written as: Since the auxiliary controller based on deep reinforcement learning needs to control the vehicle in real time according to the vehicle status, the following optimization problem is constructed: (10) Where, is the auxiliary controller action based on deep reinforcement learning, representing the auxiliary control input of the driving torque and steering angle of each wheel, represents the four-wheel independent steering and driving vehicle dynamics function, represents the auxiliary controller, Represents the upper controller, Represents the lower-level controller, Represents the lateral displacement from the differential operation and yaw angle The reference path states of the longitudinal velocity, lateral velocity, and yaw velocity are obtained; The state of the deep reinforcement learning controller is designed to be ,in Indicates the error between the current actual vehicle state and the ideal state, represents the longitudinal velocity error, represents the lateral displacement error, represents the heading angle error, represents the generalized force at the control end; The reward function of the system is designed as: (11) Where, represents the positive gain parameter of the path tracking error, A positive gain parameter representing the smoothness of continuous control, represents the limit value of tracking error, It represents the penalty factor. Indicates the changing speed of the auxiliary control variable, its purpose is to ensure a certain smoothness of the control process; Step 3: Select the twin delayed deep deterministic policy gradient algorithm to implement deep reinforcement learning; Step 4: A two-layer control architecture is used to solve the lateral and longitudinal coupling problem of the four-wheel independent steering and drive vehicle in path tracking control. The architecture consists of an upper-layer controller based on nonlinear model predictive control and a lower-layer controller based on sequential quadratic programming. The upper-layer controller calculates the current state error of the vehicle through the ideal path tracking state information and the current state of the vehicle. The optimal control increment is solved by the nonlinear model predictive controller; the lower controller solves the cost function through sequential quadratic programming and distributes the generalized control quantity of the upper controller , to calculate the lateral force and longitudinal force of each wheel at the next moment; In step 5, the vehicle is controlled through two control loops using the composite control framework.

2. The path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework according to claim 1, characterized in that: Step 1 specifically includes: the four tires are indexed as ; In the tire coordinate system, the longitudinal force, lateral force, vertical force and steering angle of each tire are expressed as and Expressed; the dynamic equation of the vehicle body is written as: (1) Where, and represents the vehicle mass and moment of inertia, Indicates the distance between the tire and the vehicle's center of gravity. and Respectively represent the longitudinal acceleration, lateral acceleration and yaw angular acceleration of the vehicle; In the design of the path tracking controller, the longitudinal and lateral forces of each tire are determined by adjusting the steering angle and wheel torque. In the tire coordinate system, the magic formula is used to express the nonlinear characteristics of the tire model: (2) Where, Indicates longitudinal force or lateral force, Indicates longitudinal slip rate or slip angle, peak coefficient , stiffness coefficient , shape coefficient and curvature coefficient is the fitting coefficient of the empirical formula; Longitudinal slip rate The calculation formula is: (3) Where, and represents the tire radius and rotational angular velocity, The actual speed at the center of the tire. The tire speed is calculated using the following formula (4) The steering angle and torque of the tire are expressed as (5) Where, is the driving torque, is the rolling friction coefficient, is the moment of inertia of the tire; Based on the above analysis, a vehicle dynamics model with seven degrees of freedom is established. By reasonably allocating input parameters, the vehicle state is effectively controlled to achieve high-performance path tracking. The control input matrix of the vehicle dynamics model is (6) In the formula, the matrix Including the driving torque and steering angle of the left front wheel, right front wheel, left rear wheel and right rear wheel of the vehicle; the vehicle state matrix is ​​defined as (7) In order to facilitate the design of auxiliary controller based on deep reinforcement learning, the dynamic equations of the four-wheel independent steering and driving vehicle are written in nonlinear matrix form (8) Where, represents the control output matrix of the two-layer controller, represents the vehicle acceleration disturbance matrix, Represents the mapping matrix between the upper controller and the lower controller control quantity, Represents the vehicle mass matrix; in order to truly reflect the impact of external disturbances on vehicle dynamics, the control end interference matrix is ​​introduced, which represents the lateral and longitudinal disturbance forces acting on each wheel (9) The external disturbance in equation (9) is transferred to the vehicle body dynamics model through the tire dynamics model to determine the vehicle body acceleration disturbance matrix .

3. The path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework according to claim 1, characterized in that: Step 3 specifically includes: before the formal training begins, the agent interacts with the environment by randomly selecting actions to accumulate enough initial experience; this step ensures that there are enough diverse samples in the experience replay pool for subsequent training; for the initial time step , the agent is in each state Next random action , record the immediate reward after executing the action and update status , and the new transfer sample Store in experience replay pool Until enough samples are stored in the experience pool; initialize the policy network , two Q networks and , and its corresponding target network 、 and , the initial parameters of the target network are the same as the current network; the experience replay pool In the first step, samples have been collected, which will be used in the subsequent training phase; From the experience replay pool Randomly sample a batch of samples from The experience replay pool breaks the temporal correlation of samples by storage and random sampling, effectively reducing the problem of policy overfitting. Then, the target policy network is used Generate next state Target action , and introduce noise Smoothing to prevent the policy network from overfitting the action (12) Using Target Q Network and Calculate the target Q value and take its minimum value to reduce the overestimation problem of the Q value (13) Where, is the target Q value, For immediate rewards; The loss function is constructed as the mean squared error between the Q-value estimate and the target Q-value: , using stochastic gradient descent to minimize the loss function of the two Q networks; through the optimization of the loss function, TD3 will gradually adjust the parameters of the Q network and , to improve the accuracy of Q value estimation; TD3 introduces a delayed policy update mechanism, that is, the policy network is updated once after every two updates of the Q network. ; The goal of the policy network is to find the action that maximizes the Q value (14) In the formula, through right Gradient propagation to iterate TD3 optimizes the policy network based on a more stable Q-value estimation by delaying the update, avoiding the policy network from relying on inaccurate Q-values ​​too early, and uses the soft update method to update the parameters of the target Q network and the target policy network, so that the two target networks are updated smoothly from the main network respectively. (15) Where, is the soft update step size; after training, the policy network Has been optimized to the optimal strategy; at this time, for a given state , directly calculate the optimal action through the policy network , the action Used to guide the agent's actual decision-making in the environment.

4. The path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework according to claim 2, wherein: The upper controller in step 4 calculates the vehicle's current state error by tracking the state information of the ideal path and the vehicle's current state. , and solve the optimal control increment through the nonlinear model predictive controller, specifically including: (16) Where, Indicates time The state error, and denote the error weight matrix and the control increment weight matrix respectively, Represents the relaxation factor to avoid the inability to find a feasible solution; the generalized control quantity is calculated by the cost function to calculate the state of the vehicle at the next moment (17) Where, 、 and They represent the generalized longitudinal force, lateral force and yaw moment of the vehicle body respectively; To balance and minimize wheel utilization, the lower controller solves the cost function through sequential quadratic programming and allocates the generalized control quantity of the upper controller , to calculate the lateral and longitudinal forces of each wheel at the next moment (18) Where, and They represent the tire utilization minimization weight coefficient and the tire utilization balance weight coefficient respectively; the lateral force and longitudinal force of each wheel are calculated as the angle and torque respectively through the tire dynamics model, and the vehicle terminal control variable at the next moment is obtained, which is then passed to the vehicle body to realize closed-loop control of path tracking.

5. The path tracking control method for a four-wheel independent steering and driving vehicle based on a composite control framework according to claim 1, characterized in that: Step 5 specifically includes: a two-layer control loop and a model-free deep reinforcement learning auxiliary control loop; the vehicle state information is transmitted to the upper-level controller to calculate the generalized force required at the next moment; at the same time, the deep reinforcement learning neural network is used as an auxiliary controller to generate auxiliary control items based on the vehicle state information and external disturbances to compensate for the upper-level control variables; the lower-level controller is responsible for distributing the generalized force transmitted by the upper-level controller and finally gives the control variables under the composite control framework.

Images