A real-time vehicle platoon automatic control method based on a zeroization neural network

Abstract

The application discloses a kind of real-time vehicle platoon automatic control method based on zero neural network, belong to the field of autonomous driving.This method will real-time vehicle platoon control problem be converted into nonlinear equation system solving problem.By using this model, the challenge of real-time vehicle platoon control can be effectively solved.In addition, this study also develops a distributed vehicle driving parameter monitoring device and controller, and constructs a new type of automatic control system.Especially, the application also proposes a predetermined time convergence zero neural network, which can not only realize the rapid convergence of the system, but also optimize the use of computing resources by adjusting time-varying parameters, significantly improving the computing efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving, and specifically to a real-time automatic vehicle platooning control method based on a null neural network. Background Technology

[0002] With technological advancements, the number of vehicles in many countries and regions has increased significantly, undoubtedly bringing a series of complex traffic challenges. Vehicle platooning control technology, as a key component of intelligent transportation systems, has attracted widespread research attention. This technology offers numerous advantages, including improved driving safety, optimized traffic flow, increased fuel efficiency, and reduced emissions. The core objective of vehicle platooning control is to ensure that each vehicle in the platoon maintains a set speed and a safe distance between vehicles. Through inter-vehicle communication technology, the perception range of vehicles is significantly expanded, enabling them to receive and exchange crucial information such as position, speed, and acceleration, thereby achieving more efficient traffic management.

[0003] To achieve vehicle platooning control, researchers have proposed various control strategies. However, the effectiveness of these methods typically relies on the assumption that the system state gradually approaches infinity over time. To address this challenge, developing strategies capable of achieving control objectives within a finite time frame has become a hot research area in academia. In conclusion, proposing an efficient and reliable control strategy for real-time vehicle platooning is crucial. Summary of the Invention

[0004] To address the aforementioned shortcomings in the existing technology, this invention provides a real-time automatic vehicle platooning control method based on a null neural network.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0006] A real-time automatic vehicle platooning control method based on a nullable neural network includes the following steps:

[0007] S1. Input the real-time vehicle platooning control problem to be solved, and design a null neural network model;

[0008] S2. Establish a mathematical model for the real-time vehicle platooning control problem based on input;

[0009] S3. An automatic control system is established based on distributed vehicle driving monitoring data and control data. The mathematical model established in S2 is used to iteratively solve the real-time vehicle formation control problem under the interference of environmental noise.

[0010] S4. Based on the results of the iterative solution in S3, control the distance, speed and acceleration of the tracked vehicle.

[0011] Furthermore, the real-time vehicle platooning control problem to be solved in S1 is expressed as:

[0012] ;

[0013] In the formula, This indicates the position where vehicle i has traveled. This represents the speed of vehicle i. Indicates the mass of vehicle i. The mechanical efficiency of the transmission system of vehicle i. This represents the actual driving or braking torque of vehicle i. It is the radius of the wheel of vehicle i; It is the overall aerodynamic drag coefficient of vehicle i; It is gravitational acceleration. It is the rolling resistance coefficient. It is the longitudinal dynamic inertial delay of vehicle i. It is the driving or braking torque required by vehicle i.

[0014] Furthermore, the nullified neural network model in S1 is represented as follows:

[0015] ;

[0016] In the formula, For a nullified neural network model; Let i be the error variable and ; All are model coefficients and among them , .

[0017] Furthermore, the establishment of the automatic control system based on distributed vehicle driving monitoring data and control data in S3 specifically includes the following steps:

[0018] S31. Monitor the status of the i-th vehicle;

[0019] S32. Quantify the performance of the monitored state using error variables;

[0020] S33. Design a new controller for the monitored vehicles.

[0021] Furthermore, the specific method for quantization performance calculation in S32 is as follows:

[0022] ;

[0023] In the formula, Let be the position error variable for the i-th vehicle. Let be the speed error variable for the i-th vehicle. Let be the acceleration error variable for the i-th vehicle. Let i be the position of the i-th vehicle. , and This indicates that vehicle i has detected the position, speed, and acceleration of the lead vehicle. Let i be the speed of vehicle i. Let i be the acceleration of vehicle i. Let i be the expected distance between vehicle i and the lead vehicle.

[0024] Furthermore, the new controller in S33 is represented as follows:

[0025] ;

[0026] In the formula, , and For fixed parameters, Let the variable be related to the residual, in and Let represent the nonlinear variables of the design with respect to vehicle i and vehicle j. These are the Laplace matrix elements related to vehicle i and vehicle j. These are the weight matrix elements of vehicle i and the virtual lead vehicle.

[0027] The present invention has the following beneficial effects:

[0028] This invention utilizes its predetermined time convergence and anti-interference capabilities, combined with the dynamics of vehicle platooning, to design an automatic control system. The advantages of this invention lie in its real-time performance, anti-interference capability, high precision, high efficiency, and low computational resource consumption. This invention solves the real-time vehicle platooning control problem using neural networks, possessing considerable novelty and practicality. Attached Figure Description

[0029] Figure 1 This is a flowchart of the real-time automatic vehicle platooning control method based on null neural networks according to the present invention.

[0030] Figure 2 This is an information diagram of the longitudinal dynamics of a vehicle according to an embodiment of the present invention.

[0031] Figure 3 This is an information topology diagram of undirected communication in vehicle platooning according to an embodiment of the present invention.

[0032] Figure 4 This is an acceleration information diagram of vehicle i in a platoon according to an embodiment of the present invention.

[0033] Figure 5 This is a speed information diagram of vehicle i in a platoon according to an embodiment of the present invention.

[0034] Figure 6 This is a diagram showing the position information of vehicle i in a platoon according to an embodiment of the present invention.

[0035] Figure 7 This is a diagram showing the spacing between vehicles in a platoon according to an embodiment of the present invention.

[0036] Figure 8 This is a diagram showing the distance between vehicle i and the virtual lead vehicle in a platoon according to an embodiment of the present invention.

[0037] Figure 9 This is a speed difference information diagram between vehicle i and the virtual lead vehicle in a platoon according to an embodiment of the present invention.

[0038] Figure 10 This is a graph showing the acceleration difference between vehicle i and the virtual lead vehicle in a platoon according to an embodiment of the present invention. Detailed Implementation

[0039] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0040] A real-time automatic vehicle platooning control method based on nullable neural networks, such as Figure 1 As shown, it includes the following steps:

[0041] S1. Input the real-time vehicle platooning control problem to be solved, and design a null neural network model;

[0042] To simplify the analysis of the control system, longitudinal tire slippage can be ignored, the vehicle body is treated as a rigid, symmetrical structure, and flexible deformation and asymmetric effects are neglected. Only the longitudinal dynamics of the vehicle are considered, reducing the complexity of the control system. Figure 2 As shown. The longitudinal dynamics of the vehicle can be expressed as:

[0043] ;

[0044] in This indicates the position where vehicle i has traveled. This represents the speed of vehicle i. Indicates the mass of vehicle i. The mechanical efficiency of the transmission system of vehicle i. This represents the actual driving or braking torque of vehicle i. It is the radius of the wheel of vehicle i; It is the overall aerodynamic drag coefficient of vehicle i; It is gravitational acceleration. It is the rolling resistance coefficient. It is the longitudinal dynamic inertial delay of vehicle i. This is the driving or braking torque required by vehicle i. The nonlinear model is then converted into a linear representation of the controller design:

[0045]

[0046] in The input is linearized, and then a new linear model of the vehicle's longitudinal dynamics can be obtained:

[0047] ;

[0048] in It is the acceleration of vehicle i. This is the input for vehicle i. This refers to the environmental factors affecting vehicle i. However, the virtual head vehicle is unaffected by external factors, so its dynamic model can be expressed as:

[0049] ;

[0050] That is, the case where i=0. , , and These represent the virtual lead car's position, velocity, acceleration, and input, respectively.

[0051] S2. Establish a mathematical model for the real-time vehicle platooning control problem based on input;

[0052] Establish a basic system of nonlinear equations, expressed as follows:

[0053]

[0054] in Indicates time, and These represent the start and end times, respectively. It is a non-linear mapping. By constructing a null neural network, we can first design... The error is Then we can obtain the model of the null neural network:

[0055] ;

[0056] in It is an activation function. The activation function designed in this invention enables the error to converge to zero rapidly within a predetermined time. The specific model is as follows: ;

[0057] in , ,and .

[0058] Then, we design a time-varying parameter that is efficient and saves computational resources. Specifically, it can be expressed as follows:

[0059] .

[0060] S3. An automatic control system is established based on distributed vehicle driving monitoring data and control data. The mathematical model established in S2 is used to iteratively solve the real-time vehicle formation control problem under the interference of environmental noise.

[0061] For monitoring the state of vehicle i, it can be represented as:

[0062]

[0063] in , and This indicates that vehicle i has detected the position, speed, and acceleration of the lead vehicle, where , and The definition is as follows:

[0064]

[0065] in This represents the symbolic function. Our expected performance for vehicle platooning control is as follows:

[0066]

[0067] in Indicates the expected distance between adjacent vehicles. This represents the predetermined time. The convergence effect at the predetermined time can be achieved through a designed activation function, as shown below:

[0068] ;

[0069] To analyze vehicle performance, a series of error variables are introduced to quantify performance:

[0070] ;

[0071] in , representing the expected distance between vehicle i and the lead vehicle.

[0072] For vehicle i, design a new controller, represented as follows:

[0073] ;

[0074] in , and All are fixed parameters. It is a positive variable parameter with respect to the residual, where Represented as .

[0075] in ;

[0076] S4. Based on the results of the iterative solution in S3, control the distance, speed and acceleration of the tracked vehicle.

[0077] Using the designed model, the iterative solution to the real-time vehicle formation control problem under environmental noise interference includes the following steps:

[0078] Provide a communication topology diagram of the vehicle platoon, such as Figure 3 As shown.

[0079] Provide relevant data on the virtual lead vehicle's movement:

[0080] Provide relevant environmental noise. Give the expected adjacent vehicle distance. Then the parameters are input into the system for solving.

[0081] Final results Figure 4 It can be seen that the vehicles in the formation are able to track the acceleration information of the virtual lead vehicle very well. Figure 5 It can be seen that the platooned vehicles can track the speed information of the virtual lead vehicle very well. Figure 6 It can be seen that the convoy vehicles can track the virtual lead vehicle's position information very well, and there will be no overtaking situations. Figure 7 and Figure 8 It can be seen that the convoy vehicles can maintain a good distance between each other, and the distance between vehicles is always at the ideal spacing of 8m, and no overtaking occurs during the tracking process. Figure 9 It can be seen that the vehicles in the formation track the speed of the virtual lead vehicle very well and can maintain the same speed as the lead vehicle. Figure 10 It can be seen that the platooned vehicles can track the acceleration of the virtual lead vehicle very well and can quickly reach a state where their acceleration is consistent with that of the lead vehicle. This example demonstrates that the neural network-based automatic control system for real-time vehicle platooning proposed in this invention has high accuracy and robustness, and its efficiency and superior performance are further confirmed by practical applications under environmental disturbances.

[0082] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0083] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0084] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0085] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

[0086] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A real-time automatic vehicle platooning control method based on a null-negative neural network, characterized in that, Includes the following steps: S1. Input the real-time vehicle platooning control problem to be solved, and design a null neural network model; S2. Establish a mathematical model for the real-time vehicle platooning control problem based on input; S3. An automatic control system is established based on distributed vehicle driving monitoring data and control data. The mathematical model established in S2 is used to iteratively solve the real-time vehicle formation control problem under environmental noise interference. The specific steps include the following: S31. Monitor the status of the i-th vehicle; S32. Quantify the performance of the monitored state using error variables. The specific method for quantifying performance is as follows: ； In the formula, Let be the position error variable for the i-th vehicle. Let be the speed error variable for the i-th vehicle. Let be the acceleration error variable for the i-th vehicle. Let i be the position of the i-th vehicle. , and This indicates that vehicle i has detected the position, speed, and acceleration of the lead vehicle. Let i be the speed of vehicle i. Let i be the acceleration of vehicle i. The expected distance between vehicle i and the lead vehicle; S33. Design a new controller for the monitored vehicles. The new controller is represented as follows: ； In the formula, , and For fixed parameters, For the variable relating to the residual, and Let represent the nonlinear variables of the design with respect to vehicle i and vehicle j. These are the Laplace matrix elements related to vehicle i and vehicle j. These are the elements of the weight matrix between vehicle i and the virtual lead vehicle; S4. Based on the results of the iterative solution in S3, control the distance, speed and acceleration of the tracked vehicle.

2. The real-time vehicle platooning automatic control method based on nullable neural networks according to claim 1, characterized in that, The real-time vehicle formation control problem to be solved in S1 is expressed as: ； In the formula, This indicates the position where vehicle i has traveled. Indicates the speed of vehicle i. Indicates the mass of vehicle i. The mechanical efficiency of the transmission system of vehicle i. This represents the actual driving or braking torque of vehicle i. It is the wheel radius of vehicle i; It is the overall aerodynamic drag coefficient of vehicle i; It is gravitational acceleration. It is the rolling resistance coefficient. It is the longitudinal dynamic inertial delay of vehicle i. It is the driving or braking torque required by vehicle i.

3. The real-time vehicle platooning automatic control method based on nullable neural networks according to claim 1, characterized in that, The nulled neural network model in S1 is represented as follows: ； In the formula, Let i be the error variable and ; All are model coefficients and among them , .

Images