A vehicle safety control algorithm based on model-free non-zero-sum game

Abstract

This invention relates to the fields of control and information technology, and particularly to a model-free non-zero-sum game-based vehicle safety control algorithm, comprising the following steps: constructing a topology diagram of inter-vehicle network communication and determining the connectivity between vehicles; obtaining the tracking error of the i-th following vehicle based on the state equations of the i-th following vehicle, the following vehicles connected to the i-th following vehicle, and the leading vehicle; defining an objective function based on the tracking error and non-zero-sum game theory, and determining the control barrier function of the control strategy network; substituting the control barrier function into the value function network to transform the constrained problem into an unconstrained problem, thereby obtaining a model-free vehicle safety control algorithm; training the control strategy network and the value function network based on the vehicle safety control algorithm to obtain the control law of the control strategy network, and controlling the i-th following vehicle according to the control law. This invention improves the robustness and adaptability of the entire system.

Description

Technical Field

[0001] This invention relates to the fields of control and information technology, and in particular to a vehicle safety control algorithm based on model-free non-zero-sum game theory. Background Technology

[0002] Vehicle control is a key technology in intelligent transportation systems. Its core function is to ensure vehicle safety—avoiding collisions while effectively reducing the distance between vehicles. This reduces air resistance and fuel consumption, thereby improving road safety and alleviating traffic congestion. With advancements in intelligent transportation systems, simply achieving basic vehicle control objectives is no longer sufficient to meet the demands of modern transportation. Current focus has shifted to optimizing vehicle platform control, including maximizing fuel efficiency, improving passenger comfort, and reducing travel time.

[0003] In recent years, with the advancement of vehicle control technology, techniques including artificial potential field methods and model predictive control have been widely applied. However, these methods often neglect the impact on other vehicles when solving the problem of safe driving of a single vehicle, and thus fail to achieve global optimum. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in related technologies. To this end, this invention provides a vehicle safety control algorithm based on model-free non-zero-sum game theory, which eliminates the need for a specific vehicle model, improves the robustness and adaptability of the entire system, and ensures the safety of vehicle operation.

[0005] This invention provides a vehicle safety control algorithm based on model-free non-zero-sum game, comprising the following steps:

[0006] Construct a topology diagram for inter-vehicle network communication, the topology diagram including a lead vehicle and... N The number of following vehicles is determined according to the topology diagram, including the leader vehicle and the following vehicles. N The connectivity relationships between the following vehicles;

[0007] According to the The state equations of the following vehicles, and the first The state equations of the leader vehicle and the following vehicles connected by the vehicle are obtained as follows: Tracking error of the following vehicle =1,2,3……, N ;

[0008] According to the The tracking error of the following vehicle and the objective function are defined using non-zero-sum game theory to determine the control barrier function of the control strategy network;

[0009] By incorporating the control barrier function as an evaluation term into the value function network, the constrained problem is transformed into an unconstrained problem, resulting in a model-free vehicle safety control algorithm.

[0010] Based on the vehicle safety control algorithm, the control strategy network and the value function network are trained to obtain the control law of the control strategy network. The control law is then used to train the first... The vehicle described above is controlled by the following vehicle.

[0011] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided to determine the leader vehicle and the following vehicles. N The connectivity between the following vehicles includes the following steps:

[0012] The topology diagram for constructing inter-vehicle network communication includes a lead vehicle and vehicles consisting of... N The undirected graph formed by the following vehicles is represented as follows: ,in, V To follow the assembly of vehicles, , E for N The communication between the following vehicles. The adjacency matrix A represents the connectivity between following vehicles. ,when >0 indicates the first The following vehicle and the first i The vehicles mentioned above are connected to each other. Indicates the relationship with the first i A group of vehicles that are in communication with each other;

[0013] The lead vehicle communicates unidirectionally with the following vehicles, where 'O' indicates the connectivity between the following vehicles and the lead vehicle. .

[0014] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the following steps are included to obtain the tracking error of the following vehicle:

[0015] According to the i The state equations of the following vehicles, and the first The state equations of the following vehicles connected to the lead vehicle and the state equation of the leader vehicle are obtained to obtain the first... i The tracking error of a following vehicle is given by the following formula:

[0016]

[0017] in, For the first Follow vehicle tThe state equation at time t, , For the first Follow vehicle t Location at any given moment For the first Follow vehicle t The speed of time, For the first Follow vehicle t acceleration at any moment For the first The differential of the state equation of the following vehicle. For the first Control inputs for following vehicles, B As the first coefficient, C As the second coefficient, For inertial time delay, ;

[0018]

[0019] in, For the first Tracking error of the following vehicle For the first j Follow vehicle t The state equation at time t, For the leader's vehicle t The state equation at time t, ,in Indicates the first The expected relative distance between the following vehicles and the leader's vehicle. , For the first The following vehicle and the first The expected relative distance of the following vehicle.

[0020] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided, and the objective function is defined by the following steps:

[0021]

[0022]

[0023] in,

[0024]

[0025]

[0026] in, For the first The cost function of a vehicle following another vehicle. For the first The control barrier function for following vehicles. For the first The control law of the vehicle-following control strategy network. In order to be with the first The control law of the control strategy network for following vehicles connected to other vehicles. In order to be with the first The tracking error of the following vehicle connected to the vehicle. This is a positive definite matrix used to adjust the degree of emphasis placed on tracking error in model-free vehicle safety control algorithms. r The function is used to determine the first The tracking error of the following vehicle and the first Following vehicles and the first The importance of the control input of the following vehicle is determined by the U-function, which limits the upper bound of the control input of the following vehicle. Its formula is as follows:

[0027]

[0028]

[0029] in, Indicates the first The upper limit of the control input for the following vehicle. s Let be the independent variable concerning the control law. Indicates the relationship with the first The upper limit of the control input for the following vehicle connected to the vehicle is specified. and It is a positive definite matrix. Restriction The upper limit of the control input for the following vehicle. Limitations and the first The upper limit of the control input for the following vehicle connected to the vehicle.

[0030] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the determination of the control barrier function includes the following steps:

[0031]

[0032]

[0033] in, for t Time of the first The following vehicle and the first Distance control parameters between following vehicles A constant that is greater than zero. No. The safe distance between following vehicles. For the first Follow vehicle t Location at any given moment Indicates the first The following vehicle and the first The expected spacing between following vehicles.

[0034] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the construction of the value function includes the following steps:

[0035] The control barrier function is added as an evaluation term to the value function network, and the value function formula of the value function network is as follows:

[0036]

[0037] in,

[0038]

[0039] in, It is a value function. z Let be the independent variable with respect to time.

[0040] According to the present invention, a model-free non-zero-sum game-based vehicle safety control algorithm is provided. Training the model-free vehicle safety control algorithm includes the following steps:

[0041] Choose the control law of the initial control strategy network and the initial network weight parameters of the value function network;

[0042] The weight parameters of the value function network are updated iteratively using the Bellman equation error, and the weight parameters and control law of the control policy network are updated iteratively in this way.

[0043] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the network weight parameters of the value function network and the control law of the control policy network are iteratively updated using the following formula:

[0044]

[0045]

[0046] in, For the first The following vehicle The value function of the next iteration For the first The following vehicle Control law of the control strategy network in the next iteration; For value function networks, The activation function of the control policy network. For value function network, the first The network weight parameters for the next iteration For the control policy network The network weight parameters for the next iteration.

[0047] The vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, which updates the weight parameters of the value function network using Bellman equation error iteration, includes the following steps:

[0048] The error of the Bellman equation is defined by the following formula:

[0049]

[0050] in, The error in the Bellman equation, For the first The difference between the control law of the given control strategy network for the following vehicle and the control law generated by the iteration;

[0051] The Bellman equation is transformed using the regression equation, resulting in the following formula:

[0052]

[0053] in,

[0054]

[0055]

[0056]

[0057] in, For the first The following vehicle and the first The interval function for each vehicle following another vehicle. For the first The following vehicle and the first The interval control error function for each following vehicle; M The moment of the first The following vehicle and the first The interval value function and interval control error function of each following vehicle are stored in a matrix, as shown in the following formula:

[0058]

[0059]

[0060] in, This is the interval control error function matrix, used to store... MThe moment of the first The following vehicle and the first The interval control error function for each following vehicle. This is an interval-valued function matrix used to store... M The moment of the first The following vehicle and the first The interval value function for each following vehicle;

[0061] The weight parameter update equation for the value function network is:

[0062] .

[0063] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, updating the weight parameters of the value function network further includes the following steps: , For the first i The following vehicle l The network weight parameters of the value function network in the second iteration and the first iteration l The minimum difference between -1 iterations is used to output the weight parameters of the value function network. Otherwise, let We continue to use the Bellman equation error iteration to update the weight parameters of the value function network.

[0064] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided, which obtains the control law of the control strategy network based on the weight parameters of the value function network.

[0065] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects:

[0066] The vehicle safety control algorithm based on model-free non-zero-sum game according to embodiments of the present invention, by adopting a model-free non-zero-sum game algorithm, eliminates the need for a specific vehicle model, enabling each vehicle to reach Nash equilibrium, balancing error and energy consumption, and ensuring the safety of vehicle operation.

[0067] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0069] Figure 1 This is one of the flowcharts of the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention.

[0070] Figure 2 This is a topology diagram of inter-vehicle network communication based on a model-free non-zero-sum game-based vehicle safety control algorithm provided by the present invention.

[0071] Figure 3 This is a graph showing the distance between the third and fourth following vehicles over time in the vehicle safety control algorithm based on model-free non-zero-sum game provided by this invention.

[0072] Figure 4 The following diagram shows the variation of the tracking error of the following vehicle over time in the vehicle safety control algorithm based on model-free non-zero-sum game provided by this invention. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but cannot be used to limit the scope of this invention.

[0074] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0075] Non-zero-sum game theory offers a new perspective on solving the problem of single-vehicle safe driving, which often overlooks the impact on other vehicles. In the non-zero-sum game framework, each vehicle is treated as an independent entity striving to maximize its own predetermined goals, transforming the centralized energy optimization control problem into a non-zero-sum game model. Based on this, a control barrier function is introduced to ensure driving safety. This method transforms the fully state-constrained system into an unconstrained multi-vehicle safe cooperative system and ensures the overall stability of the system by establishing a Nash equilibrium.

[0076] To address the aforementioned issues, a model-free non-zero-sum game-based vehicle safety control algorithm is proposed. The basic idea is to design a control barrier function to balance the system's safety and optimality based on the vehicle topology communication structure, establish a new reinforcement learning method for non-zero-sum games and derive it into a model-free algorithm to obtain the corresponding control inputs for each vehicle.

[0077] Figure 1 This is one of the flowcharts of the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention.

[0078] This invention provides a vehicle safety control algorithm based on model-free non-zero-sum game theory, such as... Figure 1 As shown, it includes the following steps:

[0079] Construct a topology diagram for inter-vehicle network communication, the topology diagram including a lead vehicle and... N The number of following vehicles is determined according to the topology diagram, including the leader vehicle and the following vehicles. N The connectivity relationships between the following vehicles;

[0080] According to the The state equations of the following vehicles, and the first The state equations of the leader vehicle and the following vehicles connected by the vehicle are obtained as follows: The tracking error of the following vehicle. =1,2,3……, N ;

[0081] According to the The tracking error of the following vehicle and the objective function defined by non-zero-sum game theory are used to determine the control barrier function of the control policy (Actor) network;

[0082] By incorporating the control barrier function as an evaluation term into the Critic network, the constrained problem is transformed into an unconstrained problem, resulting in a model-free vehicle safety control algorithm.

[0083] Based on the vehicle safety control algorithm, the control strategy network and the value function network (Actor-Critic network) are trained to obtain the control law of the control strategy network. The control law is then used to train the first... The vehicle described above is controlled by the following vehicle.

[0084] In this embodiment, by employing a model-free non-zero-sum game algorithm, without the need for a specific vehicle model, each vehicle can reach Nash equilibrium, balancing error and energy consumption, and ensuring the safety of vehicle operation.

[0085] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided to determine the leader vehicle and the following vehicles. N The connectivity between the following vehicles includes the following steps:

[0086] The topology diagram for constructing inter-vehicle network communication includes a lead vehicle and vehicles consisting of... N The undirected graph formed by the following vehicles is represented as follows: ,in, V To follow the assembly of vehicles, , E for N The communication between the following vehicles. The adjacency matrix A represents the connectivity between following vehicles. ,when >0 indicates the first The following vehicle and the first i The vehicles mentioned above are connected to each other. Indicates the relationship with the first i The set of vehicles that are in communication with the following vehicle;

[0087] The lead vehicle communicates unidirectionally with the following vehicles, where O represents the connectivity between the following vehicles and the lead vehicle. ,when Time indicates the first i The following vehicles can receive information from the leading vehicle.

[0088] In this embodiment, the communication network topology diagram has low structural requirements, does not require a specific vehicle model, and only needs to communicate with local neighbor agents. It has low communication costs, strong adaptability, and is easy to implement.

[0089] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the following steps are included to obtain the tracking error of the following vehicle:

[0090] The tracking error of the vehicle is obtained from the state equation of the following vehicle, and its formula is as follows:

[0091]

[0092]

[0093] in, For the first Follow vehicle t The state equation at time t, , For the first Follow vehicle t Location at any given moment For the first Follow vehicle t The speed of time, For the first Follow vehicle t acceleration at any moment For the first The differential of the state equation of the following vehicle. For the first Control inputs for following vehicles, B As the first coefficient, C As the second coefficient, For inertial time delay, ;

[0094]

[0095] in, For the first Tracking error of the following vehicle For the first j Follow vehicle t The state equation at time t, For the leader's vehicle t The state equation at time t, ,in Indicates the first The expected relative distance between the following vehicles and the leader's vehicle. , For the first The following vehicle and the first The expected relative distance of the following vehicle.

[0096] In this embodiment, by calculating the tracking error and relative distance of the following vehicle, the safety of vehicle operation can be guaranteed.

[0097] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided, and the objective function is defined by the following steps:

[0098]

[0099]

[0100] in,

[0101]

[0102]

[0103] in,

[0104] in, For the first The cost function of a vehicle following another vehicle. For the first The control barrier function for following vehicles. For the first The control law of the vehicle-following control strategy network. In order to be with the first The control law of the control strategy network for following vehicles connected to other vehicles. In order to be with the first The tracking error of the following vehicle connected to the vehicle. This is a positive definite matrix used to adjust the degree of emphasis placed on tracking error in model-free vehicle safety control algorithms. r The function is used to determine the first The tracking error of the following vehicle and the first Following vehicles and the first The importance of the control input of the following vehicle is determined by the U-function, which limits the upper bound of the control input of the following vehicle. Its formula is as follows:

[0105]

[0106]

[0107] in, Indicates the first The upper limit of the control input for the following vehicle. s Let be the independent variable concerning the control law. Indicates the relationship with the first The upper limit of the control input for the following vehicle connected to the vehicle is specified. and It is a positive definite matrix. Restriction The upper limit of the control input for the following vehicle. Limitations and the first The upper limit of the control input for the following vehicle connected to the vehicle.

[0108] In this embodiment, the robustness and adaptability of the entire system are improved by employing a model-free non-zero-sum game algorithm.

[0109] According to some embodiments of the present invention, a non-zero-sum game is a cooperative game in which the sum of the gains or losses of each party is not zero.

[0110] According to some embodiments of the present invention, a strategy combination reaches Nash equilibrium when each player's equilibrium strategy is to maximize their expected payoff, and all other players follow the same strategy.

[0111] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the determination of the control barrier function includes the following steps:

[0112]

[0113]

[0114] in, for t Time of the first The following vehicle and the first Distance control parameters between following vehicles A constant that is greater than zero. No. The safe distance between following vehicles. For the first Follow vehicle t Location at any given moment Indicates the first The following vehicle and the first The expected spacing between following vehicles.

[0115] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the construction of the value function includes the following steps:

[0116] The control barrier function is added as an evaluation term to the value function network, and the value function formula of the value function network is as follows:

[0117]

[0118] in,

[0119]

[0120] in, It is a value function. z Let be the independent variable with respect to time.

[0121] In this embodiment, the safety and optimality of the system are balanced by designing a control barrier function.

[0122] This invention provides a vehicle safety control algorithm based on model-free non-zero-sum game theory.

[0123] Training a model-free vehicle safety control algorithm includes the following steps:

[0124] Choose the control law of the initial control strategy network and the initial network weight parameters of the value function network;

[0125] The weight parameters of the value function network are updated iteratively using the Bellman equation error, and the weight parameters and control law of the control policy network are updated iteratively in this way.

[0126] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, the network weight parameters of the value function network and the control law of the control policy network are iteratively updated using the following formula:

[0127]

[0128]

[0129] in, For the first The following vehicle The value function of the next iteration For the first The following vehicle Control law of the control strategy network in the next iteration; For value function networks, The activation function of the control policy network. For value function network, the first The network weight parameters for the next iteration For the control policy network The network weight parameters for the next iteration.

[0130] The vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, which updates the weight parameters of the value function network using Bellman equation error iteration, includes the following steps:

[0131] The error of the Bellman equation is defined by the following formula:

[0132]

[0133] in, The error in the Bellman equation, For the first The difference between the control law of the given control strategy network for the following vehicle and the control law generated by the iteration;

[0134] The Bellman equation is transformed using the regression equation, resulting in the following formula:

[0135]

[0136] in,

[0137]

[0138]

[0139]

[0140] Will M The moment of the first The following vehicle and the first The interval value function and interval control error function of each following vehicle are stored in a matrix, as shown in the following formula:

[0141]

[0142]

[0143] in, This is the interval control error function matrix, used to store... M The moment of the first The following vehicle and the first The interval control error function for each following vehicle. This is an interval-valued function matrix used to store... M The moment of the first The following vehicle and the first The interval value function for each following vehicle;

[0144] The weight parameter update equation for the value function network is:

[0145]

[0146] According to the vehicle safety control algorithm based on model-free non-zero-sum game provided by the present invention, updating the weight parameters of the value function network further includes the following steps: , For the first i The following vehicle l The network weight parameters of the value function network in the second iteration and the first iteration l The minimum difference between -1 iterations is used to output the weight parameters of the value function network. Otherwise, let We continue to use the Bellman equation error iteration to update the weight parameters of the value function network.

[0147] According to some preferred embodiments of the present invention, =0.1.

[0148] According to the present invention, a vehicle safety control algorithm based on model-free non-zero-sum game is provided, which obtains the control law of the control strategy network based on the weight parameters of the value function network.

[0149] The technical solution of the present invention will be further explained below with reference to a specific embodiment. It should be noted that the specific embodiment is only for the purpose of enabling those skilled in the art to better understand the technical solution of the present invention, and should not be regarded as an unreasonable limitation on the scope of protection of the present invention.

[0150] Example 1

[0151] Figure 2 This is a topology diagram of inter-vehicle network communication based on a model-free non-zero-sum game-based vehicle safety control algorithm provided by the present invention. Figure 3 This is a graph showing the distance between the third and fourth following vehicles over time in the vehicle safety control algorithm based on model-free non-zero-sum game provided by this invention. Figure 4 The following diagram shows the variation of the tracking error of the following vehicle over time in the vehicle safety control algorithm based on model-free non-zero-sum game provided by this invention.

[0152] In the simulation experiment, the topology diagram of the inter-vehicle network communication is assumed to be as follows: Figure 2 As shown. For each vehicle, (identity matrix) , , The control inputs of each vehicle satisfy Safe following distance The ideal vehicle spacing is 7. For the first... Vehicles following each other, defining the tracking error. That is, the error in position, velocity, and acceleration, defined as... Based on the topology diagram of the inter-vehicle network communication, the activation function of the Critic network for each vehicle can be set. , , , , Its initial weight parameters Each term is 1. Similarly, the activation function of the Actor network for each car can be set. , , , , Its initial weight parameters Each item is 1.

[0153] like Figure 3As shown, without considering the control barrier function (Without CBF), the distance between the third and fourth following vehicles may be too small to be dangerous (Risky region). However, with the control barrier function (With CBF), the distance between the third and fourth following vehicles can be strictly guaranteed to be within the safe region, demonstrating that the model can improve the robustness and adaptability of the entire system.

[0154] like Figure 4 As shown, after learning is completed, the tracking errors of the first, second, third, fourth, and fifth following vehicles gradually approach 0, and the closer they are to the leader vehicle, the easier it is for them to stabilize.

[0155] An embodiment of the present invention provides a vehicle safety control algorithm based on model-free non-zero-sum game theory, which has the following technical advantages compared to the prior art:

[0156] (1) No specific vehicle model is required, which improves the robustness and adaptability of the entire system.

[0157] (2) The network communication topology has low structural requirements, strong applicability, and is easy to implement.

[0158] (3) Each following vehicle only needs to communicate with local neighbor following vehicles, resulting in low communication costs and ensuring the safety of vehicle operation.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle safety control algorithm based on model-free non-zero-sum game theory, characterized in that, Includes the following steps: Construct a topology diagram for inter-vehicle network communication, the topology diagram including a lead vehicle and... N The number of following vehicles is determined according to the topology diagram, including the leader vehicle and the following vehicles. N The connectivity relationships between the following vehicles; According to the The state equations of the following vehicles, and the first The state equations of the leader vehicle and the following vehicles connected by the aforementioned following vehicles are obtained as follows: The tracking error of the following vehicle. =1,2,3……, N ; According to the The tracking error of the following vehicle and the objective function are defined using non-zero-sum game theory to determine the control barrier function of the control strategy network; Defining the objective function includes the following steps: in, in, For the first The cost function of a vehicle following another vehicle. For the first The control barrier function for following vehicles. For the first Tracking error of the following vehicle For the first The control law of the vehicle-following control strategy network. In order to be with the first The control law of the control strategy network for following vehicles connected to other vehicles. In order to be with the first The tracking error of the following vehicle connected to the vehicle. This is a positive definite matrix used to adjust the degree of emphasis placed on tracking error in model-free vehicle safety control algorithms. For the first Follow vehicle t Location at any given moment For the first Follow vehicle t Location at any given moment r The function is used to determine the first The tracking error of the following vehicle and the first Following vehicles and the first The importance attached to the control input of the following vehicle. Restriction The upper limit of the control input for the following vehicle. Limitations and the first The upper bound of the control input of the following vehicle is defined by the U function, which is used to limit the upper bound of the control input of the following vehicle. Its formula is as follows: in, Indicates the first The upper limit of the control input for the following vehicle. s Let be the independent variable concerning the control law. Indicates the relationship with the first The upper limit of the control input for the following vehicle connected to the vehicle is specified. and It is a positive definite matrix; Determining the control barrier function includes the following steps: in, for t Time of the first The following vehicle and the first Distance control parameters between following vehicles A constant that is greater than zero. No. The safe distance between following vehicles. Indicates the first The following vehicle and the first The expected spacing between following vehicles; By substituting the control barrier function as an evaluation term into the value function network, the constrained problem is transformed into an unconstrained problem, resulting in a model-free vehicle safety control algorithm. Based on the vehicle safety control algorithm, the control strategy network and the value function network are trained to obtain the control law of the control strategy network. The control law is then used to train the first... The vehicle being followed is controlled; The control law for iteratively updating the network weight parameters and control policy network of the value function network is adopted using the following formula: in, For the first The following vehicle The value function of the next iteration For the first The following vehicle Control law of the control strategy network in the next iteration; For value function networks, The activation function of the control policy network. For value function network, the first The network weight parameters for the next iteration For the control policy network The network weight parameters for the next iteration.

2. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 1, characterized in that, Determine the leader vehicle and the follower vehicle, and N The connectivity between the following vehicles includes the following steps: The topology diagram for constructing inter-vehicle network communication includes a lead vehicle and vehicles consisting of... N The undirected graph formed by the following vehicles is represented as follows: ,in, V To follow the assembly of vehicles, , E for N The communication between the following vehicles. The adjacency matrix A represents the connectivity between following vehicles. ,when >0 indicates the first The following vehicle and the first i The vehicles mentioned above are connected to each other. Indicates the relationship with the first i The set of vehicles that are in communication with the following vehicle; The lead vehicle communicates unidirectionally with the following vehicles, where 'O' indicates the connectivity between the following vehicles and the lead vehicle. .

3. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 2, characterized in that, Obtaining the tracking error of the following vehicle involves the following steps: According to the i The state equations of the following vehicles, and the first The state equations of the following vehicles connected to the lead vehicle and the state equation of the leader vehicle are obtained to obtain the first... i The tracking error of a following vehicle is given by the following formula: in, For the first Follow vehicle t The state equation at time 10:00 , For the first Follow vehicle t Location at any given moment For the first Follow vehicle t The speed of time, For the first Follow vehicle t acceleration at any moment For the first The differential of the state equation of the following vehicle. For the first Control inputs for following vehicles, B As the first coefficient, C As the second coefficient, For inertial delay, ; in, For the first Tracking error of the following vehicle For the first j Follow vehicle t The state equation at time t, For the leader's vehicle t The state equation at time t, ,in Indicates the first The expected relative distance between the following vehicles and the leader's vehicle. , For the first The following vehicle and the first The expected relative distance of the following vehicle.

4. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 3, characterized in that, Constructing a value function involves the following steps: The control barrier function is added as an evaluation term to the value function network, and the value function formula of the value function network is as follows: in, in, It is a value function. z Let be the independent variable with respect to time.

5. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 4, characterized in that, Training a model-free vehicle safety control algorithm includes the following steps: Choose the control law of the initial control strategy network and the initial network weight parameters of the value function network; The weight parameters of the value function network are updated iteratively using the Bellman equation error, and the weight parameters and control law of the control policy network are updated iteratively in this way.

6. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 5, characterized in that, Updating the weight parameters of a value function network using the Bellman equation error iteration involves the following steps: The error of the Bellman equation is defined by the following formula: in, The error in the Bellman equation is... For the first The difference between the control law of the given control strategy network for the following vehicle and the control law generated by the iteration; The Bellman equation is transformed using the regression equation, resulting in the following formula: in, in, For the first The following vehicle and the first The interval function for each vehicle following another vehicle. For the first The following vehicle and the first The interval control error function for each following vehicle; M The moment of the first The following vehicle and the first The interval value function and interval control error function of each following vehicle are stored in a matrix, as shown in the following formula: in, This is the interval control error function matrix, used to store... M The moment of the first The following vehicle and the first The interval control error function for each following vehicle. This is an interval-valued function matrix used to store... M The moment of the first The following vehicle and the first The interval value function for each following vehicle; The weight parameter update equation for the value function network is: 。 7. The vehicle safety control algorithm based on model-free non-zero-sum game as described in claim 6, characterized in that, Updating the weight parameters of the value function network also includes the following steps: , For the first i The following vehicle l The network weight parameters of the value function network in the second iteration and the first iteration l The minimum difference between -1 iterations is used to output the weight parameters of the value function network. Otherwise, let We continue to use the Bellman equation error iteration to update the weight parameters of the value function network.

Images