Vehicle queue optimal control method and system based on function deviation triggering mechanism

Abstract

The invention relates to the technical field of vehicle control, in particular to a vehicle queue optimal control method and system based on a function deviation triggering mechanism, and the method comprises the steps: 1, constructing a piloting dynamics model of a piloting vehicle and a following dynamics model of a following vehicle; 2, defining a local tracking error of the ith following vehicle, defining control input of the following dynamical model as a function of the local tracking error according to an optimal control target, further defining an optimal value function, and determining optimal control input according to the optimal value function; 3, obtaining an optimal value function estimation value through an adaptive commentator neural network; 4, designing a function deviation triggering mechanism according to the optimal value function estimation value and the local tracking error so as to dynamically adjust the optimal control input of the following dynamical model; and 5, setting a communication association relationship and an initial state of the leading vehicle and the following vehicle to obtain a queue control result. According to the invention, efficient cooperative driving in a complex dynamic environment is realized.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to a vehicle queue optimal control method and system based on a function deviation trigger mechanism. Background Art

[0002] With traffic congestion and energy consumption becoming increasingly prominent, vehicle platoon control, as a key technology in intelligent transportation systems, has attracted widespread attention. By coordinating the distance, speed, and acceleration between vehicles, vehicle platoon control enables safe and efficient cooperative driving, effectively alleviating traffic pressure and reducing fuel consumption. While traditional vehicle platoon control strategies, such as leader-follower control, model predictive control, and sliding mode control, have achieved some success, these methods typically rely on precise dynamic models and require complex online optimization, which limits their flexibility and robustness in dynamic and uncertain traffic environments.

[0003] Reinforcement learning, a method that can obtain optimal control strategies without requiring a precise system dynamics model, has attracted increasing attention for solving complex platoon control problems. Meanwhile, advances in wireless communication technologies have enabled vehicles to share real-time information through vehicle-to-vehicle communication. However, limited bandwidth and the high demand for real-time communication pose challenges to the implementation of platoon control. Event-triggered mechanisms can reduce redundant data transmission and alleviate communication overhead by only communicating when preset trigger conditions are met. However, the integration of existing event-triggered mechanisms with reinforcement learning frameworks within the framework of vehicle platoon control requires further research. Furthermore, traditional static event-triggered mechanisms lack adaptability, making it difficult to dynamically adjust trigger conditions during the learning process. Dynamic event-triggered mechanisms, on the other hand, struggle to pre-design trigger conditions based on specified performance, leading to reliability issues in practical applications. Summary of the Invention

[0004] The present invention aims to address at least one of the technical problems existing in the related art. To this end, it provides a method and system for optimal control of vehicle platoons based on a function deviation triggering mechanism. This method addresses the existing issues of relying on precise models, high communication overhead, and insufficient adaptability of event-triggered mechanisms, thereby enabling efficient cooperative driving in complex dynamic environments.

[0005] The present invention provides a vehicle platoon optimal control method based on a function deviation trigger mechanism, comprising the following steps: Step 1: Construct the pilot dynamics model of the pilot vehicle and Following dynamics model of a following vehicle; Step 2: Define the local tracking error of the i-th following vehicle. According to the optimal control objective, define the control input of the following dynamics model as a function of the local tracking error, and then define the optimal value function. Then, determine the optimal control input according to the optimal value function. Step 3: Obtain the optimal value function estimate through the adaptive critic neural network; Step 4: Design a function deviation trigger mechanism based on the optimal value function estimate and the local tracking error to dynamically adjust the optimal control input of the following dynamics model; Step 5: Set the communication relationship and initial state between the lead vehicle and the following vehicle, and execute steps 1 to 4 to obtain the queue control result.

[0006] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step 1 specifically includes: The vehicle platoon consists of a lead vehicle and The index of the pilot car is defined as 0, and the index of the following car is defined as ; The pilot dynamics model of the pilot vehicle is: , in, for The state vector of the pilot car at any moment, , for The position of the pilot car at all times, for The speed of the pilot car at all times, for The acceleration of the pilot car at all times, is the system matrix; The following dynamics model of the following vehicle is: , in, for The state vector of the i-th following vehicle at time, , for The position of the i-th following car at time, for The speed of the i-th following car at time, for The acceleration of the i-th following car at time, is the system matrix, for The control input of the i-th following vehicle at time.

[0007] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step 2 specifically includes: The local tracking error of the i-th following vehicle is defined as ,but in, is the local tracking error of the i-th following vehicle, is the communication relationship between the i-th following car and the j-th following car. If the i-th following car and the j-th following car can communicate directly, then ,otherwise ; is the communication relationship between the i-th following car and the pilot car. If the i-th following car can directly receive the pilot car information, then ,otherwise ; , is the expected relative distance between the i-th following car and the leading car, is the expected relative distance between the i-th following car and the j-th following car, is the desired distance between the i-th following vehicle and the leading vehicle, The desired distance between the i-th following vehicle and the j-th following vehicle; for The state vector of the jth following vehicle at time, , for The position of the jth following car at time, for The speed of the jth following car at time, for The acceleration of the jth following vehicle at time.

[0008] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step 2 further includes the following steps: The control input is defined as a function of the local tracking error, then The control input of the i-th following vehicle at time for ; Introducing time-domain quadratic performance indicators in, is the performance index of the i-th following vehicle, is the weighting matrix of the local tracking error, is a positive definite matrix, is the weighted scalar of the control input, ; Defining a value function for in, is the integral variable in the integral operation; The optimal value function is .

[0009] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step 2 further includes the following steps: According to the optimality principle, the Hamiltonian function is defined as for in, is the control input set of the neighbors of the i-th following vehicle, is the set of local tracking errors of the neighbors of the i-th following vehicle, for right The gradient, is the number of neighboring following vehicles associated with the i-th following vehicle, is the control input of the jth following vehicle; By minimizing the Hamiltonian function, the optimal control input can be obtained as .

[0010] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step 2 further includes the following steps: Adopting the event trigger control strategy, the event trigger time sequence is defined as , at each time the i-th following car The time when the trigger control input is updated , the local tracking error of the i-th following vehicle is sent to the controller to update the optimal control input; In the interval between two adjacent triggers, the control input remains unchanged based on the zero-order hold mechanism, that is, , in, is the i-th following car The time when the trigger control input is updated, is the i-th following car The time when the trigger control input is updated.

[0011] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step three includes: Design an adaptive critic neural network, and use the approximation ability of the adaptive critic neural network to express the optimal value function as in, is the ideal weight vector, is the activation function of the neural network, is the approximate error; Approximation based on adaptive critic neural network , which is in the form of in, is the optimal value function estimate, is the estimated weight vector of the critic network; According to the estimated value of the optimal value function, the optimal control input is in, is the estimated value of the optimal control input, is the gradient of the activation function with respect to the local tracking error; Under the event trigger mechanism, the control input remains constant between consecutive trigger moments, so in, is the estimated value of the optimal control input under the event-triggered mechanism; Based on the integral reinforcement learning method, for a given time period, the time difference error is defined as in, is the time difference vector of the activation function, for Performance cost within a time period, is the time difference error, is the time period; definition For the critic weight estimation error, define the instantaneous error function , using the gradient descent method, the update law of the estimated weight vector of the critic network is: in, is the learning rate, , is an auxiliary variable.

[0012] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step four specifically includes: Design function deviation trigger mechanism in, is the gradient of the optimal value function estimate with respect to the local tracking error, is the preset attenuation function, is the first parameter of the trigger mechanism, , The second parameter of the trigger mechanism, , is a preset threshold parameter related to vehicle i, representing the upper bound of the optimal value function relative to the local tracking error. express The minimum eigenvalue of .

[0013] A further improvement of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention is that step five specifically includes: Set the communication relationship, initial state, and time period between the lead vehicle and the following vehicle , the weighted matrix of the local tracking error , a weighted scalar of the control input , learning rate , the first parameter of the trigger mechanism , trigger mechanism second parameter , preset threshold parameters related to vehicle i , run steps 1 to 4 to obtain the queue control result.

[0014] The present invention also provides a vehicle platoon optimal control system based on a function deviation trigger mechanism, the vehicle platoon optimal control system is used to execute the vehicle platoon optimal control method described above, and the vehicle platoon optimal control system includes: Model building module, used to build the pilot dynamics model of the pilot car and Following dynamics model of a following vehicle; The control input determination module is used to define the local tracking error of the following vehicle. According to the optimal control objective, the control input of the following dynamics model is defined as a function of the local tracking error. Then, the optimal value function is defined to determine the optimal control input. The value function estimation module is used to obtain the optimal value function estimation value through the adaptive critic neural network; The control input adjustment module is used to design a function deviation trigger mechanism based on the optimal value function estimate and local tracking error, and dynamically adjust the optimal control input of the following dynamics model; The platoon control operation module is used to set the communication relationship and initial state between the lead vehicle and the following vehicle, and sequentially calls the functions of the model construction module, control input determination module, value function estimation module, and control input adjustment module to obtain the platoon control results.

[0015] The present invention dynamically adjusts the trigger condition according to the deviation between the optimal value function and the preset attenuation function, and combines integral reinforcement learning and adaptive critic neural network to ensure system stability and control accuracy while reducing communication overhead.

[0016] This invention employs reinforcement learning to learn optimal control strategies without requiring a precise system model, adapting to the uncertainties of dynamic traffic environments. A trigger mechanism based on function deviation dynamically adjusts trigger conditions based on the deviation between the current optimal value function and a preset attenuation function, achieving a trade-off between communication frequency and stability. Control inputs are designed based on optimal control theory, ensuring that they theoretically minimize preset performance indicators. This provides a globally optimal theoretical basis for cooperative driving in vehicle platoons.

[0017] The present invention supports a variety of communication topologies. For the two-way pilot-leader following, two-way pilot-tail following, two-way lead following, and two-way dual pilot-leader following topologies, the designed methods can ensure that the queuing system has good performance stability.

[0018] This invention primarily focuses on designing a trigger mechanism based on function deviation, aiming to solve the problem of vehicle platoon control based on reinforcement learning. Its core approach is to introduce a dynamic deviation term between the optimal value function and a preset attenuation function, using this as the basis for adjusting the trigger conditions, thereby constructing a performance-aware adaptive trigger mechanism. On the one hand, compared to traditional static event triggering mechanisms, this mechanism exhibits greater flexibility, effectively reducing unnecessary control updates and communication events, thereby reducing communication overhead. On the other hand, it overcomes the theoretical bottleneck of traditional dynamic event triggering mechanisms under preset performance constraints, significantly improving the utilization efficiency of communication resources while ensuring the system's preset performance. Ultimately, this mechanism provides a highly efficient collaborative control solution for vehicle platoon systems that both guarantees preset performance and possesses dynamic adaptability.

[0019] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is a flow chart of the vehicle queue optimal control method based on the function deviation trigger mechanism provided by the present invention.

[0022] Figure 2 is a schematic diagram of different communication topologies.

[0023] Figure 3This is a schematic diagram of a simulation experiment of the vehicle queue optimal control method based on the function deviation trigger mechanism provided by the present invention. Figure 1 .

[0024] Figure 4 This is a schematic diagram of a simulation experiment of the vehicle queue optimal control method based on the function deviation trigger mechanism provided by the present invention. Figure 2 .

[0025] Figure 5 This is a schematic diagram of a simulation experiment of the vehicle queue optimal control method based on the function deviation trigger mechanism provided by the present invention. Figure 3 . DETAILED DESCRIPTION

[0026] To make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. Obviously, the embodiments described are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0027] The following combination Figure 1 The vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention includes the following steps: Step 1: Construct the pilot dynamics model of the pilot vehicle and Following dynamics model of a following vehicle; Step 2: Define the local tracking error of the i-th following vehicle. According to the optimal control objective, define the control input of the following dynamics model as a function of the local tracking error, and then define the optimal value function. Then, determine the optimal control input according to the optimal value function. Step 3: Obtain the optimal value function estimate through the adaptive critic neural network; Step 4: Design a function deviation trigger mechanism based on the optimal value function estimate and the local tracking error to dynamically adjust the optimal control input of the following dynamics model; Step 5: Set the communication relationship and initial state between the lead vehicle and the following vehicle, and execute steps 1 to 4 to obtain the queue control result.

[0028] Preferably, by constructing a pilot dynamics model of the pilot vehicle and The following dynamics model of each follower vehicle accurately simulates the motion state of the vehicle platoon, providing a foundation for subsequent control strategy design, which helps improve the overall motion accuracy and coordination of the vehicle platoon. Defining the control input of the following dynamics model as a function of the local tracking error ensures that the following vehicle closely tracks the trajectory of the lead vehicle, reducing error accumulation and improving platoon stability. Furthermore, determining the optimal control input based on the optimal value function further optimizes the control strategy, achieving more efficient energy utilization and lowering energy consumption. Obtaining an estimate of the optimal value function through an adaptive critic neural network allows for real-time adjustment of the control strategy to adapt to varying traffic conditions and vehicle states, improving control flexibility and adaptability and helping the platoon maintain stable operation in complex and changing traffic environments. A function deviation trigger mechanism, designed based on the optimal value function estimate and local tracking error, dynamically adjusts the optimal control input of the following dynamics model, promptly correcting deviations and ensuring that the vehicle platoon consistently maintains the desired trajectory, thereby improving the safety and reliability of the platoon. By setting the communication relationship and initial states between the lead and follower vehicles and running the entire control process, accurate platoon control results can be obtained.

[0029] In a preferred embodiment of the vehicle platoon optimal control method based on the function deviation trigger mechanism of the present invention, step 1 specifically includes: The vehicle platoon consists of a lead vehicle and The index of the pilot car is defined as 0, and the index of the following car is defined as ; The pilot dynamics model of the pilot vehicle is: , in, for The state vector of the pilot car at any moment, , for The position of the pilot car at all times, for The speed of the pilot car at all times, for The acceleration of the pilot car at all times, is the system matrix; The following dynamics model of the following vehicle is: , in, for The state vector of the i-th following vehicle at time, , for The position of the i-th following car at time, for The speed of the i-th following car at time, for The acceleration of the i-th following car at time, is the system matrix, for The control input of the i-th following vehicle at time.

[0030] Furthermore, step 2 specifically includes: The local tracking error of the i-th following vehicle is defined as ,but in, is the local tracking error of the i-th following vehicle, is the communication relationship between the i-th following car and the j-th following car. If the i-th following car and the j-th following car can communicate directly, then ,otherwise ; is the communication relationship between the i-th following car and the pilot car. If the i-th following car can directly receive the pilot car information, then ,otherwise ; , is the expected relative distance between the i-th following car and the leading car, is the expected relative distance between the i-th following car and the j-th following car, is the desired distance between the i-th following vehicle and the leading vehicle, The desired distance between the i-th following vehicle and the j-th following vehicle; for The state vector of the jth following vehicle at time, , for The position of the jth following car at time, for The speed of the jth following car at time, for The acceleration of the jth following vehicle at time.

[0031] Preferably, Indicates that the state deviation of the i-th following vehicle and other following vehicles is arrive The traversal accumulation of .

[0032] Furthermore, step 2 also includes the following steps: The control input is defined as a function of the local tracking error, then The control input of the i-th following vehicle at time for ; Introducing time-domain quadratic performance indicators in, is the performance index of the i-th following vehicle, is the weighting matrix of the local tracking error, is a positive definite matrix, Satisfy the positive definite matrix requirement, thus ensuring that the performance index is strictly positive definite for the error penalty, avoiding the performance index being zero when the error is non-zero, and ensuring that the control strategy effectively guides the system toward stability; is the weighted scalar of the control input, ; In order to calculate the optimal control input for each following vehicle, for any given admissible control input, we define the value function for in, is the integral variable in the integral operation; The optimal value function is .

[0033] Furthermore, step 2 also includes the following steps: According to the optimality principle, the Hamiltonian function is defined as for in, is the control input set of the neighbors of the i-th following vehicle, is the set of local tracking errors of the neighbors of the i-th following vehicle, for right The gradient, is the number of neighboring following vehicles associated with the i-th following vehicle, is the control input of the jth following vehicle; By minimizing the Hamiltonian function, the optimal control input can be obtained as .

[0034] Furthermore, step 2 also includes the following steps: In order to reduce the computation and communication burden, an event-triggered control strategy is adopted, and the event triggering time sequence is defined as , at each time the i-th following car The time when the trigger control input is updated , the local tracking error of the i-th following vehicle is sent to the controller to update the optimal control input; In the interval between two adjacent triggers, the control input remains unchanged based on the zero-order hold mechanism, that is, , in, is the i-th following car The time when the trigger control input is updated, is the i-th following car The time when the trigger control input is updated.

[0035] Therefore, the control objective can be defined as: Given a vehicle platoon control system that satisfies the leading dynamics model and the following dynamics model in step 1, design a suitable event triggering mechanism to determine the trigger sequence And the corresponding events trigger the optimal control input to reduce the communication frequency while ensuring system performance.

[0036] Preferably, the triggered event can be the norm of the error or the comparison result thereof with a preset threshold. When the local tracking error exceeds the threshold, the event is triggered, and this moment is recorded as the event triggering moment. Subsequently, the local tracking error information of the following vehicle is transmitted to the controller to update the control input.

[0037] Preferably, the triggering time may also be a time variable, and an optimal triggering time sequence may be determined through algorithm optimization to achieve a balance between communication frequency and system performance.

[0038] Preferably, the trigger condition is dynamically adjusted based on the deviation between the current optimal value function and a preset attenuation function, achieving a trade-off between communication frequency and stability. Control inputs are designed based on optimal control theory to ensure that they theoretically minimize preset performance indicators, providing a globally optimal theoretical basis for cooperative driving in vehicle platoons.

[0039] Furthermore, step three includes: Design an adaptive critic neural network, and use the approximation ability of the adaptive critic neural network to express the optimal value function as in, is the ideal weight vector, is the activation function of the neural network, is the approximate error; Approximation based on adaptive critic neural network , which is in the form of in, is the estimated value of the optimal value function, is the estimated weight vector of the critic network; According to the estimated value of the optimal value function, the optimal control input is in, is the estimated value of the optimal control input, is the gradient of the activation function with respect to the local tracking error; Under the event trigger mechanism, the control input remains constant between consecutive trigger moments, so in, is the estimated value of the optimal control input under the event trigger mechanism; under the event trigger mechanism, when When the system meets the event-triggered control strategy, the control input is updated and a new control input is generated. , and upon reaching Before the triggering moment, the control input remains unchanged. During this time period, the control input is continuously based on State deviation Output constant action; Based on the integral reinforcement learning method, for a given time period, the time difference error is defined as in, is the time difference vector of the activation function, for Performance cost within a time period, is the time difference error, is the time period; definition For the critic weight estimation error, define the instantaneous error function , using the gradient descent method, the update law of the estimated weight vector of the critic network is: in, is the learning rate, , is an auxiliary variable.

[0040] Advantageously, the adaptive critic neural network possesses powerful approximation capabilities, capable of expressing complex optimal value functions in a concise mathematical form. This simplifies the design process of vehicle platoon control systems, reduces the requirements for system modeling accuracy, and improves the adaptability and robustness of the control system. The adaptive critic neural network enables precise estimation of the optimal control input. Based on the estimated optimal value function and taking into account the gradient of the activation function with respect to the local tracking error, this ensures the accuracy and effectiveness of the control input, helping to improve the tracking accuracy and stability of the vehicle platoon, and enhancing overall control performance.

[0041] Preferably, based on the integral reinforcement learning method, the estimated weight vector of the adaptive critic neural network can be continuously updated to minimize the immediate error function. The gradient descent method is used to ensure the convergence and stability of learning. Through continuous learning and optimization, the control system can gradually approach the optimal solution, achieving more efficient and intelligent vehicle platoon control.

[0042] Furthermore, step four specifically includes: Design function deviation trigger mechanism in, is the gradient of the optimal value function estimate with respect to the local tracking error, is the preset attenuation function, is the first parameter of the trigger mechanism, , The second parameter of the trigger mechanism, , is a preset threshold parameter related to vehicle i, representing the upper bound of the optimal value function relative to the local tracking error. express The minimum eigenvalue of is the function deviation term, that is, the deviation between the preset attenuation function and the estimated value of the optimal value function. Under the function deviation event trigger mechanism, the system only triggers when the trigger condition is met: When the trigger control input is updated; compared with the traditional static event trigger mechanism, the function deviation event trigger mechanism can dynamically adjust the trigger condition according to the function deviation term, showing greater flexibility and ensuring that the system meets the stability requirements.

[0043] Preferably, the function deviation trigger mechanism helps to reduce energy consumption. By precisely controlling the relative position and speed between vehicles, unnecessary acceleration and braking operations are reduced, thereby reducing overall energy consumption, improving energy utilization efficiency, and enhancing the robustness of the system. In the face of complex traffic environments and changeable vehicle dynamics characteristics, the function deviation trigger mechanism can maintain a stable control effect, ensuring that the vehicle queue can maintain the optimal state under various conditions.

[0044] Furthermore, step five specifically includes: setting the communication relationship, initial state, time period of the pilot vehicle and the following vehicle , the weighted matrix of the local tracking error , a weighted scalar of the control input , learning rate , the first parameter of the trigger mechanism , trigger mechanism second parameter , preset threshold parameters related to vehicle i , run steps 1 to 4 to obtain the queue control result.

[0045] The present invention also provides a vehicle platoon optimal control system based on a function deviation trigger mechanism, the vehicle platoon optimal control system is used to execute the vehicle platoon optimal control method described above, and the vehicle platoon optimal control system includes: Model building module, used to build the pilot dynamics model of the pilot car and Following dynamics model of a following vehicle; The control input determination module is used to define the local tracking error of the following vehicle. According to the optimal control objective, the control input of the following dynamics model is defined as a function of the local tracking error. Then, the optimal value function is defined to determine the optimal control input. The value function estimation module is used to obtain the optimal value function estimation value through the adaptive critic neural network; The control input adjustment module is used to design a function deviation trigger mechanism based on the optimal value function estimate and local tracking error, and dynamically adjust the optimal control input of the following dynamics model; The platoon control operation module is used to set the communication relationship and initial state between the lead vehicle and the following vehicle, and sequentially calls the functions of the model construction module, control input determination module, value function estimation module, and control input adjustment module to obtain the platoon control results.

[0046] In a specific implementation case, a simulation experiment is conducted using the above vehicle platoon optimal control method. A total of five vehicles are set up, including one pilot vehicle and four follower vehicles. The communication relationship of the vehicle platoon is set. The communication topology of the vehicle platoon is as follows: Figure 2 As shown, (a) Bidirectional pilot-front-wheel drive following (BLPF), (b) Bidirectional pilot-tail vehicle following (BLTF), (c) Bidirectional front-wheel drive following (BPF), (d) Bidirectional dual pilot-front-wheel drive following (BTLPF), the initial state of the vehicle is set to Setting parameters The expected relative distance between adjacent vehicles is 7 meters, and the trigger frequency is defined as the number of triggers / the number of simulation steps. The simulation results are as follows: Figure 3 、 Figure 4 and Figure 5 As shown, Figure 3 、 Figure 4 and Figure 5 They represent the local tracking error, velocity, and acceleration of the following vehicle under the BLPF topology, respectively. The triggering frequencies of each following vehicle are 6.64%, 9.62%, 4.80%, 40.44%, and 6.22%, respectively. Table 1 shows the average absolute spacing error under different topologies.

[0047] Table 1 Experiments show that the following vehicle can maintain the desired relative position with the lead vehicle, and the speed and acceleration remain synchronized. Under the designed function deviation trigger mechanism, the communication frequency between vehicles is greatly reduced.

[0048] Preferably, compared with the existing technology, the specific beneficial technical effects of the present invention are: using a reinforcement learning method, the optimal control strategy can be learned without a precise system model, adapting to the uncertainty in the dynamic traffic environment; utilizing a trigger mechanism based on function deviation, the trigger condition is dynamically adjusted according to the deviation between the current optimal value function and the preset attenuation function, achieving a trade-off between communication frequency and stability; designing the control input based on optimal control theory to ensure that the control input can theoretically minimize the preset performance index, providing a global optimal theoretical basis for the cooperative driving of vehicle platoons; supporting multiple communication topologies, and for the topologies of two-way pilot-leader follower, two-way pilot-tail follower, two-way lead follower, and two-way dual pilot-leader follower, the designed methods can all ensure good performance stability of the platoon system.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A vehicle platoon optimal control method based on a function deviation trigger mechanism, characterized in that: The steps include: Step 1: Construct the pilot dynamics model of the pilot vehicle and Following dynamics model of a following vehicle; The vehicle platoon consists of a lead vehicle and The index of the pilot car is defined as 0, and the index of the following car is defined as ; The pilot dynamics model of the pilot vehicle is: , in, for The state vector of the pilot car at any moment, , for The position of the pilot car at all times, for The speed of the pilot car at all times, for The acceleration of the pilot car at all times, is the system matrix; The following dynamics model of the following vehicle is: , in, for The state vector of the i-th following vehicle at time, , for The position of the i-th following car at time, for The speed of the i-th following car at time, for The acceleration of the i-th following car at time, is the system matrix, for The control input of the i-th following vehicle at time; Step 2: Define the local tracking error of the i-th following vehicle. According to the optimal control objective, define the control input of the following dynamics model as a function of the local tracking error, and then define the optimal value function. Then, determine the optimal control input according to the optimal value function. Step 3: Obtain the optimal value function estimate through the adaptive critic neural network; Step 4: Design a function deviation trigger mechanism based on the optimal value function estimate and the local tracking error to dynamically adjust the optimal control input of the following dynamics model; Step 5: Set the communication relationship and initial state between the lead vehicle and the following vehicle, and execute steps 1 to 4 to obtain the queue control result.

2. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 1 is characterized in that: Step 2 specifically includes: The local tracking error of the i-th following vehicle is defined as ,but in, is the local tracking error of the i-th following vehicle, is the communication relationship between the i-th following car and the j-th following car. If the i-th following car and the j-th following car can communicate directly, then ,otherwise ; is the communication relationship between the i-th following car and the pilot car. If the i-th following car can directly receive the pilot car information, then ,otherwise ; , is the expected relative distance between the i-th following car and the leading car, is the expected relative distance between the i-th following car and the j-th following car, is the desired distance between the i-th following vehicle and the leading vehicle, The desired distance between the i-th following vehicle and the j-th following vehicle; for The state vector of the jth following vehicle at time, , for The position of the jth following car at time, for The speed of the jth following car at time, for The acceleration of the jth following vehicle at time.

3. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 2 is characterized in that: Step 2 also includes the following steps: The control input is defined as a function of the local tracking error, then The control input of the i-th following vehicle at time for ; Introducing time-domain quadratic performance indicators in, is the performance index of the i-th following vehicle, is the weighting matrix of the local tracking error, is a positive definite matrix, is the weighted scalar of the control input, ; Defining a value function for in, is the integral variable in the integral operation; The optimal value function is .

4. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 3 is characterized in that: Step 2 also includes the following steps: According to the optimality principle, the Hamiltonian function is defined as for in, is the control input set of the neighbors of the i-th following vehicle, is the set of local tracking errors of the neighbors of the i-th following vehicle, for right The gradient, is the number of neighboring following vehicles associated with the i-th following vehicle, is the control input of the jth following vehicle; By minimizing the Hamiltonian function, the optimal control input can be obtained as .

5. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 4 is characterized in that: Step 2 also includes the following steps: Adopting the event trigger control strategy, the event trigger time sequence is defined as , each time the i-th following car The time when the trigger control input is updated , the local tracking error of the i-th following vehicle is sent to the controller to update the optimal control input; In the interval between two adjacent triggers, the control input remains unchanged based on the zero-order hold mechanism, that is, , in, is the i-th following car The time when the trigger control input is updated, is the i-th following car The time when the trigger control input is updated.

6. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 5 is characterized in that: Step three includes: Design an adaptive critic neural network, and use the approximation ability of the adaptive critic neural network to express the optimal value function as in, is the ideal weight vector, is the activation function of the neural network, is the approximate error; Approximation based on adaptive critic neural network , which is of the form in, is the estimated value of the optimal value function, is the estimated weight vector of the critic network; According to the estimated value of the optimal value function, the optimal control input is in, is the estimated value of the optimal control input, is the gradient of the activation function with respect to the local tracking error; Under the event trigger mechanism, the control input remains constant between consecutive trigger moments, so in, is the estimated value of the optimal control input under the event-triggered mechanism, is the i-th following car The time when the secondary event is triggered; Based on the integral reinforcement learning method, for a given time period, the time difference error is defined as in, is the time difference vector of the activation function, for Performance cost within a time period, is the time difference error, is the time period; definition For the critic weight estimation error, define the instantaneous error function , using the gradient descent method, the update law of the estimated weight vector of the critic network is: in, is the learning rate, , is an auxiliary variable.

7. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 6 is characterized in that: Step 4 specifically includes: Design function deviation trigger mechanism in, is the gradient of the optimal value function estimate with respect to the local tracking error, is the preset attenuation function, is the first parameter of the trigger mechanism, , The second parameter of the trigger mechanism, , is a preset threshold parameter related to vehicle i, representing the upper bound of the optimal value function relative to the local tracking error. for The minimum eigenvalue of .

8. The vehicle platoon optimal control method based on the function deviation trigger mechanism according to claim 7 is characterized in that: Step 5 specifically includes: Set the communication relationship, initial state, and time period between the lead vehicle and the following vehicle , the weighted matrix of the local tracking error , a weighted scalar of the control input , learning rate , the first parameter of the trigger mechanism , trigger mechanism second parameter , preset threshold parameters related to vehicle i , run steps 1 to 4 to obtain the queue control result.

9. The optimal control system for vehicle platoons based on a function deviation trigger mechanism is characterized by: The vehicle platoon optimal control system is used to execute the vehicle platoon optimal control method according to any one of claims 1 to 8, and the vehicle platoon optimal control system includes: Model building module, used to build the pilot dynamics model of the pilot car and Following dynamics model of a following vehicle; The control input determination module is used to define the local tracking error of the following vehicle. According to the optimal control objective, the control input of the following dynamics model is defined as a function of the local tracking error. Then, the optimal value function is defined to determine the optimal control input. The value function estimation module is used to obtain the optimal value function estimation value through the adaptive critic neural network; The control input adjustment module is used to design a function deviation trigger mechanism based on the optimal value function estimate and local tracking error, and dynamically adjust the optimal control input of the following dynamics model; The platoon control operation module is used to set the communication relationship and initial state between the lead vehicle and the following vehicle, and sequentially calls the functions of the model construction module, control input determination module, value function estimation module, and control input adjustment module to obtain the platoon control results.

Images