Tssznn-dc1 / 2-based interconnected autonomous vehicle platoon dynamic constraint control method

Abstract

The application discloses a kind of based on TSSZNN-DC1 / 2 interconnection automatic driving vehicle formation dynamic constraint control method, it includes the following steps: constructing the dynamic constraint control system of interconnection automatic driving vehicle formation;TSSZNN-DC1 discrete model and TSSZNN-DC2 discrete model are constructed, realize the more optimal dynamic constraint control of interconnection automatic driving vehicle formation under certain circumstances.This application designs and uses time-varying bounded attenuation coefficient based on topological feedback in TSSZNN-DC1 discrete model and TSSZNN-DC2 discrete model design, so that the model has faster convergence speed while reducing the system burden caused by too large data, simultaneously design the strong bounded constraint function based on error norm gain, for dynamic constraint control of vehicle formation, more in line with the formation situation under actual traffic scene, and still adopt the step-by-step gain coefficient based on error, enhanced the anti-interference performance of model, the model has the advantages such as fast convergence speed, high precision, low latency, strong anti-interference ability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of interconnected automatic driving vehicle formation control, in particular to an interconnected automatic driving vehicle formation dynamic constraint control method based on TSSZNN-DC1 / 2 BACKGROUND

[0002] In the traditional research of interconnected automatic driving vehicle formation, a controller is mainly designed according to the needs and combined with a specific observer to realize the control of the position, speed and acceleration of each vehicle in the interconnected automatic driving formation. There are many optimized solutions for the formation control of a few vehicles on a straight road, but if the number of vehicles in the vehicle formation design is large, the controller of the vehicle will not only encounter external disturbances of the road when solving the equations of the system, but also be disturbed by the noise generated by the sensor itself and other problems. This is not desirable when implementing the control of interconnected automatic driving vehicle formation. Some documents give the design of a nonlinear manifold and a distributed observer to solve the disturbance problem, so as to obtain an exact solution. Then this method mainly uses fixed parameters, which is a relatively inefficient process in the experiment, and the system convergence speed is slow and there is a time delay problem.

[0003] In order to solve this problem, the existing technical documents propose a method of using a time-varying parameter for the observer. After introducing the time-varying parameter, the convergence time of the system is shortened, but since the time-varying increasing parameter is introduced into the observer, as time increases, the time-varying parameter will tend to infinity, causing the observer equipment to be damaged, increasing the calculation cost, making it difficult to solve in real time, and the acceleration size will be too large, which has a bad influence on the comfort experience of passengers, and the acceleration size does not conform to the actual road conditions and vehicle performance. SUMMARY

[0004] In view of the above problems in the prior art, the interconnected automatic driving vehicle formation dynamic constraint control method based on TSSZNN-DC1 / 2 provided by the present application, wherein TSSZNN-DC1 / 2 represents TSSZNN-DC1 discrete model or TSSZNN-DC2 discrete model, and the specific meaning is: containing a time-varying bounded attenuation coefficient based on topological feedback, a strongly bounded dynamic constraint function based on error norm gain, and a zeroization neural network dynamic constraint discrete model 1 or 2 based on error step gain coefficient. The problem that the prior art is difficult to stably and effectively control the interconnected automatic driving vehicle formation and the problem of difficult to realize effective acceleration dynamic constraint are solved.

[0005] In order to achieve the above application purposes, the technical scheme adopted by the present application is:

[0006] The application provides an interconnection automatic driving vehicle platoon dynamic constraint control method based on a TSSZNN-DC1 / 2, which comprises the following steps.

[0007] S1, setting a longitudinal dynamics system of the interconnection automatic driving vehicle;

[0008] S2, processing a longitudinal dynamics system equation of the interconnection automatic driving vehicle by using a feedback linearization technology and adding disturbance to obtain a linear longitudinal dynamics system of the interconnection automatic driving vehicle;

[0009] S3, constructing a longitudinal dynamics system of a virtual signal interconnection automatic driving vehicle based on the obtained linear longitudinal dynamics system of the interconnection automatic driving vehicle;

[0010] S4, respectively discretizing the linear longitudinal dynamics system equation processed in S2 and the longitudinal dynamics system equation of the virtual signal interconnection automatic driving vehicle obtained in S3 by using a forward Euler discretization method, and correspondingly obtaining a discretized vehicle platoon longitudinal dynamics system equation and a discretized virtual longitudinal dynamics system;

[0011] S5, constructing an interconnection automatic driving vehicle platoon dynamic constraint control system equation and an error function thereof according to the discretized vehicle platoon longitudinal dynamics system equation and the discretized virtual longitudinal dynamics system equation;

[0012] S6, respectively constructing a TSSZNN-DC1 discrete model and a TSSZNN-DC2 discrete model according to the discretized vehicle platoon longitudinal dynamics system equation and the discretized virtual longitudinal dynamics system and for undirected communication topology and directed communication topology;

[0013] S7, observing the interconnection automatic driving vehicle in different topologies by using the constructed TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model, performing real-time dynamic constraint control, and completing the interconnection automatic driving vehicle platoon dynamic constraint control.

[0014] The beneficial effects of the present application are: the present application designs and adopts a new distributed observer, a controller, a nonlinear manifold, a time-varying bounded attenuation coefficient based on topological feedback, a strongly bounded dynamic constraint function based on error norm gain, and a step-by-step gain coefficient based on error in the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model, the new distributed observer enables the present application to track the virtual signal vehicle in real time, the controller enables the present application to converge in a limited time, the nonlinear manifold enables the model to converge under noise interference, the time-varying bounded attenuation coefficient based on topological feedback prevents the observer parameters of the present application from overfitting, the strongly bounded dynamic constraint function based on error norm gain enables the present application to effectively dynamically constrain the acceleration of the vehicle, optimizes the comfort experience of passengers in vehicle platooning, and is more in line with the platooning situation in actual traffic scenarios, the step-by-step gain coefficient based on error enhances the anti-interference ability of the controller, and has the advantages of high precision and low time delay. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a flowchart of the method;

[0016] Figure 2 is an undirected communication topology graph based on the method provided in the embodiments of the present application;

[0017] Figure 3 is a front vehicle following mode graph in a directed communication topology based on the method provided in the embodiments of the present application;

[0018] Figure 4 is a virtual signal-front vehicle following mode graph in a directed communication topology based on the method provided in the embodiments of the present application;

[0019] Figure 5 is a comparison graph of the time-varying bounded attenuation coefficient method based on topological feedback and the exponential function coefficient, fixed coefficient and first function coefficient for solving the convergence precision of the interconnection autonomous vehicle platoon control system to achieve the ideal distance between vehicles under the condition of bounded Gaussian white noise interference and road interference in an undirected communication topology platoon;

[0020] Figure 6 is a comparison graph of the time-varying bounded attenuation coefficient method based on topological feedback and the exponential function coefficient, fixed coefficient and first function coefficient for solving the convergence precision of the interconnection autonomous vehicle platoon control system to achieve the ideal distance between vehicles under the condition of bounded Gaussian white noise interference and road interference in a front vehicle following mode in a directed communication topology platoon;

[0021] Figure 7The convergence accuracy comparison chart of the topological feedback-based time-varying bounded attenuation coefficient method and the exponential function coefficient, the fixed coefficient and the first function coefficient for solving the interconnection autonomous vehicle platoon control system to achieve the ideal distance between vehicles under the condition of bounded Gaussian white noise interference and road interference in the virtual signal-vehicle following mode of the directed communication topology formation;

[0022] Figure 8 The comparison chart of the actual position, speed and acceleration of each vehicle of the interconnection autonomous vehicle platoon dynamic constraint control system equation and the expected value under the condition of bounded Gaussian white noise and road interference, which is solved by the method under the condition of undirected communication topology connection;

[0023] Figure 9 The error chart of the actual position, speed and acceleration of each vehicle of the interconnection autonomous vehicle platoon control system and the expected value under the condition of bounded Gaussian white noise and road interference, which is solved by the method under the condition of undirected communication topology connection; wherein (a) is the position difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle; (b) is the position difference between adjacent interconnection autonomous vehicles; (c) is the speed difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle; (d) is the acceleration difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle;

[0024] Figure 10 The comparison chart of the actual position, speed and acceleration of each vehicle of the interconnection autonomous vehicle platoon dynamic constraint control system equation and the expected value under the condition of bounded Gaussian white noise and road interference, which is solved by the method under the condition of directed communication topology front-following mode connection;

[0025] Figure 11 The error chart of the actual position, speed and acceleration of each vehicle of the interconnection autonomous vehicle platoon control system and the expected value under the condition of bounded Gaussian white noise and road interference, which is solved by the method under the condition of directed communication topology front-following mode connection; wherein (a) is the position difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle; (b) is the position difference between adjacent interconnection autonomous vehicles; (c) is the speed difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle; (d) is the acceleration difference between the interconnection autonomous vehicle and the virtual interconnection autonomous vehicle;

[0026] Figure 12 The comparison chart of the actual position, speed and acceleration of each vehicle of the interconnection autonomous vehicle platoon dynamic constraint control system equation and the expected value under the condition of bounded Gaussian white noise and road interference, which is solved by the method under the condition of directed communication topology virtual signal-vehicle following mode connection;

[0027] Figure 13In the presence of bounded Gaussian white noise and road disturbance, the method solves the error map of the actual position, velocity and acceleration of each vehicle of the interconnected autonomous vehicle formation control system in the virtual signal-vehicle following mode connection state of the directed communication topology; wherein (a) is the position difference between the interconnected autonomous vehicle and the virtual interconnected autonomous vehicle; (b) is the position difference between adjacent interconnected autonomous vehicles; (c) is the speed difference between the interconnected autonomous vehicle and the virtual interconnected autonomous vehicle; (d) is the acceleration difference between the interconnected autonomous vehicle and the virtual interconnected autonomous vehicle;

[0028] Figure 14 In the presence of bounded Gaussian white noise and road disturbance, the method solves the actual acceleration of each vehicle of the interconnected autonomous vehicle formation dynamic constraint control system equation in the same case as other conditions, without the strong bounded constraint function term, and compares the actual acceleration solved in the case without the strong bounded constraint function term;

[0029] Figure 15 In the presence of bounded Gaussian white noise and road disturbance, the method solves the actual acceleration of each vehicle of the interconnected autonomous vehicle formation dynamic constraint control system equation in the directed communication topology in the vehicle following mode connection state, and compares the actual acceleration solved in the case without the strong bounded constraint function term in the same case as other conditions;

[0030] Figure 16 In the presence of bounded Gaussian white noise and road disturbance, the method solves the actual acceleration of each vehicle of the interconnected autonomous vehicle formation dynamic constraint control system equation in the virtual signal-vehicle following mode connection state of the directed communication topology, and compares the actual acceleration solved in the case without the strong bounded constraint function term in the same case as other conditions. DETAILED DESCRIPTION

[0031] The specific embodiments of the present application are described below to facilitate the understanding of the core content of the present application by those skilled in the art, but it should be clear that the present application is not limited to the scope of the specific embodiments, and according to the idea of the present application, there will be changes in the specific embodiments and application scope, and for ordinary skilled in the art, as long as various changes are within the scope of the appended claims and the scope of the present application, these changes are obvious, and all the inventions and creations using the concept of the present application are included in the protection.

[0032] As Figure 1 shown, the interconnected autonomous vehicle formation dynamic constraint control method based on TSSZNN-DC1 / 2 includes the following steps:

[0033] S1, setting the longitudinal dynamics system of the interconnected autonomous vehicle;

[0034] In this embodiment, the expression of the longitudinal dynamics system equation of the vehicle is:

[0035]

[0036] wherein and respectively represent the position, speed, actual driving / braking torque and desired driving / braking torque of the nth interconnected autonomous vehicle; respectively represent the first order derivative of and respectively represent the total external force, mass, mechanical efficiency of the powertrain system, tire radius, combined aerodynamic drag coefficient, gravity, rolling resistance coefficient and inertia delay of the longitudinal dynamics of the nth interconnected autonomous vehicle; t represents time.

[0037] S2, using feedback linearization technology to process the longitudinal dynamics system equation of the interconnected autonomous vehicle and adding disturbance, obtaining the linear longitudinal dynamics system of the interconnected autonomous vehicle;

[0038] The specific method of using feedback linearization technology to process the longitudinal dynamics system equation of the interconnected autonomous vehicle and adding disturbance in step S2 includes the following sub-steps:

[0039] S21, constructing the expression of feedback linearization technology:

[0040]

[0041] wherein respectively represent the gravity acceleration and the control input of the nth interconnected autonomous vehicle after linearization;

[0042] S22, applying the linearization expression to the longitudinal dynamics system equation of the interconnected autonomous vehicle, obtaining the linearized longitudinal dynamics system equation of the vehicle, the expression of which is:

[0043]

[0044] wherein is the acceleration of the nth interconnected autonomous vehicle; is the first order derivative of

[0045] ​​​​​​​S23, the disturbance is added to the longitudinal dynamics system equation of the linearized vehicle to obtain a longitudinal dynamics system equation after adding the disturbance, and the expression is:

[0046]

[0047] wherein is the disturbance of the i-th interconnected autonomous vehicle;

[0048] S24, let , to obtain a rewritten longitudinal dynamics system equation, and the expression is:

[0049]

[0050] wherein is the disturbance of the i-th interconnected autonomous vehicle after the new control input of the i-th interconnected autonomous vehicle; is the dynamic constraint controller to be designed for the i-th interconnected autonomous vehicle. S3, based on the obtained linear longitudinal dynamics system of the interconnected autonomous vehicle, a longitudinal dynamics system of a virtual signal interconnected autonomous vehicle is constructed;

[0051] The expression of the longitudinal dynamics system equation of the virtual signal interconnected autonomous vehicle in step S3 is:

[0052]

[0053] wherein

[0054] , , , and respectively represent the position, speed, acceleration and control input of the virtual signal interconnected autonomous vehicle; , and are the first derivatives of , and .

[0055] S4, the linear longitudinal dynamics system equation processed in S2 and the longitudinal dynamics system equation of the virtual signal interconnected autonomous vehicle obtained in S3 are respectively discretized by using the forward Euler discretization method, and the discretized vehicle platoon longitudinal dynamics system equation and the discretized virtual longitudinal dynamics system are obtained.

[0056] The expression of the discretized vehicle platoon longitudinal dynamics system equation in step S4 is:

[0057] ​​

[0058] in , , , and They represent the first Connected autonomous vehicles Always , , , and The sampled values, , and Represent , and The first derivative;

[0059] The discretized virtual longitudinal dynamic system equations in step S4 are expressed as follows:

[0060]

[0061] in , , and These respectively represent virtual signal interconnected autonomous vehicles in Always , , and The sampled values, , and Represent , and The first derivative.

[0062] S5. Construct the dynamic constraint control system equations and error functions for the interconnected autonomous vehicle platooning based on the discretized longitudinal dynamic system equations of the platoon and the discretized virtual longitudinal dynamic system equations.

[0063] The expression for the equations of the connected autonomous vehicle platooning dynamic constraint control system in step S5 is as follows:

[0064]

[0065] in The first column in this column vector The element is the first Connected autonomous vehicles Always The sampled values; the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for the j-th element of the column vector is the sampled value of the j-th interconnected autonomous vehicle at time t for , and are the first derivatives of , and respectively; denotes the transpose of a matrix; is the total number of interconnected autonomous vehicles; for the disturbance its expression is:

[0066]

[0067] where denotes the minimum of the two, denotes the maximum of the two, is the upper bound of acceleration, is the lower bound of acceleration, denotes the Gaussian white noise, represents the bounded Gaussian white noise;

[0068] The expression of the error function of the interconnected autonomous vehicle formation dynamic constraint control system equation is:

[0069]

[0070] where , and are column vectors composed of the observation values of each interconnected autonomous vehicle for , and of the virtual signal interconnected autonomous vehicle respectively; is a column vector consisting of the designed distance between each connected autonomous vehicle and the virtual signal connected autonomous vehicle, is the designed distance between adjacent vehicles; and are column vectors consisting of the error values of each connected autonomous vehicle in position, velocity and acceleration, respectively, relative to the virtual signal connected autonomous vehicle.

[0071] S6, constructing a TSSZNN-DC1 discrete model and a TSSZNN-DC2 discrete model according to the discretized platoon longitudinal dynamics system equation and the discretized virtual longitudinal dynamics system, and for the undirected communication topology and the directed communication topology, respectively;

[0072] The TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model in step S6 both include a distributed observer, a dynamic constraint controller, a nonlinear manifold, a time-varying bounded attenuation coefficient based on topology feedback, a strongly bounded dynamic constraint function based on error norm gain, and a step-by-step gain coefficient based on error; wherein the expression of the distributed observer of the TSSZNN-DC1 discrete model for the undirected communication topology is:

[0073]

[0074] wherein and represent the values of and at the th step, is the time step; is the enhanced Laplacian matrix of the connected autonomous vehicles, wherein is the Laplacian matrix of the undirected graph, wherein is the adjacency matrix of the undirected graph, when the th vehicle keeps communication with the th vehicle, otherwise, both are 0, is the diagonal matrix of , wherein n represents the number of vehicles, , …, ; is the communication link matrix of the connected autonomous vehicles and the virtual signal connected autonomous vehicle, when the th signal connected autonomous vehicle can communicate with the virtual signal connected autonomous vehicle,​​​​​​ , otherwise ; , , ; is a sign function, whose expression is:

[0075]

[0076] is a time-varying bounded attenuation coefficient based on topological feedback, whose expression is: where , is a constant to be set, denotes the minimum eigenvalue of the obtained matrix, denotes the time of sampling value;

[0077] Dynamic constraint controller of TSSZNN-DC1 discrete model whose expression is:

[0078]

[0079] where , , , is a constant to be set; , , is a step gain coefficient based on error, whose expression is: , , , , , is a constant to be set, denotes the two-norm; The whole is expressed as a strongly bounded dynamic constraint function based on error norm gain, where the error norm gain whose expression is: , is expressed as a one-norm, where is expressed as a settable boundary, , where: , is a settable upper bound of acceleration, is a settable lower bound of acceleration; is a nonlinear manifold, whose expression is:

[0080] The expression of the distributed observer of the TSSZNN-DC2 discrete model for the directed communication topology in step S6 is:

[0081]

[0082] where , and represent , and the value of the i-th step, respectively, is the time step; where is the Laplacian matrix of the directed communication topology graph of the interconnected autonomous vehicles, there are two topologies as follows:

[0083]

[0084]

[0085] is represented as the front-following mode, is represented as the virtual signal-front-following mode; is represented as a constant to be set; , , ; is a sign function, whose expression is:

[0086]

[0087] is a time-varying bounded attenuation coefficient based on topology feedback, whose expression is: where , are constants to be set, , , , , represents the minimum eigenvalue of the obtained matrix, and n represents the number of vehicles, represents the time of sampling;

[0088] Dynamic constraint controller of TSSZNN-DC2 discrete model whose expression is:

[0089]

[0090] where , , , are constants to be set; , , is an error-based step gain coefficient, whose expression is: , 、 , , , is a constant to be set, denotes a two-norm; is denoted as a strongly bounded dynamic constraint function based on error norm gain, where the error norm gain is expressed as: , is denoted as a one-norm, where is denoted as a settable boundary, , where: , is a settable upper bound of acceleration, is a settable lower bound of acceleration; is a nonlinear manifold, which is expressed as:

[0091] .

[0092] S7, by constructing the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model, observing the interconnected autonomous vehicles in different topological conditions, and performing real-time dynamic constraint control, the formation dynamic constraint control of the interconnected autonomous vehicles is completed.

[0093] The expression of the first derivative of the position, velocity and acceleration of the interconnected autonomous vehicle in the equation of the formation dynamic constraint control system of the interconnected autonomous vehicles in step S7 is solved by the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model:

[0094]

[0095] The specific method for solving the equation of the formation dynamic constraint control system of the interconnected autonomous vehicles by the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model is:

[0096] Let

[0097] where ; denotes the iterative update value of ; is the first derivative of ; is the acceleration of the virtual signal interconnected autonomous vehicle; is the speed of the virtual signal interconnected autonomous vehicle; is denoted as , is denoted as , Represented as ;

[0098] By By substituting these values ​​into the TSSZNN-DC1 or TSSZNN-DC2 discrete model, the position, velocity, and acceleration of each connected autonomous vehicle in either the undirected or directed communication topology can be obtained.

[0099] In practical implementation, the acceleration of the virtual signal interconnected autonomous vehicle is expressed as:

[0100]

[0101] In one embodiment of the present invention, the three communication topologies of the vehicle are as follows: Figure 2 , Figure 3 and Figure 4 To verify the effectiveness of this invention, the method was compared with methods using other function coefficients (e.g., exponential function coefficients, fixed coefficients, and linear function coefficients), and the results are as follows. Figure 5 , Figure 6 , Figure 7 As shown. From Figures 5-7 It can be intuitively seen that the convergence speed of this method to the desired inter-vehicle distance is faster than that of other methods using function coefficients; in terms of constraint capability, this method is superior to other solutions that do not use strong bounded constraint functions in solving the constraint capability of the dynamic constraint control system for connected autonomous vehicle platooning.

[0102] Figure 8 , Figure 9 , Figure 10 , Figure 11 , Figure 12 , Figure 13 , Figure 14 , Figure 15 and Figure 16 This paper presents a comparison of the actual position, velocity, and acceleration of each vehicle with the expected values, as well as an error plot, for solving the equations of the dynamic constraint control system for connected autonomous vehicle platooning under bounded Gaussian white noise and road disturbances using the proposed method. Figures 8-16 As can be seen, under suitable parameters, this method can achieve an error of 10% between the actual position, velocity, and acceleration of each vehicle and the expected value. -4 Furthermore, the method converges within a finite time. Therefore, it is evident that this method offers advantages of high accuracy and low latency in solving the equations of dynamic constraint control systems for connected autonomous vehicle platooning.

Claims

1. A dynamic constraint control method for platooning interconnected automated vehicles based on TSSZNN-DC1 / 2, characterized in that, Includes the following steps: S1. Establish the longitudinal dynamics system for connected autonomous vehicles; S2. The longitudinal dynamics system equations of the connected autonomous vehicle are processed and perturbed using feedback linearization technology to obtain the linear longitudinal dynamics system of the connected autonomous vehicle. S3. Construct the longitudinal dynamics system of the virtual signal-connected autonomous vehicle based on the obtained linear longitudinal dynamics system of the connected autonomous vehicle; S4. The linear longitudinal dynamics system equations processed in S2 and the longitudinal dynamics system equations of the virtual signal interconnected autonomous vehicle obtained in S3 are discretized using the forward Euler discretization method, respectively, to obtain the discretized fleet longitudinal dynamics system equations and the discretized virtual longitudinal dynamics system. S5. Construct the dynamic constraint control system equations and error functions for the interconnected autonomous vehicle platooning based on the discretized longitudinal dynamic system equations of the platoon and the discretized virtual longitudinal dynamic system equations. S6. Based on the discretized longitudinal dynamics system equations of the convoy and the discretized virtual longitudinal dynamics system, and for undirected communication topology and directed communication topology respectively, construct the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model. S7. By constructing the TSSZNN-DC1 discrete model and the TSSZNN-DC2 discrete model, we observe interconnected autonomous vehicles with different topological conditions and perform real-time dynamic constraint control to complete the dynamic constraint control of interconnected autonomous vehicle platooning. In step S6, both the TSSZNN-DC1 and TSSZNN-DC2 discrete models include a distributed observer, a dynamic constraint controller, a nonlinear manifold, a time-varying bounded decay coefficient based on topological feedback, a strongly bounded dynamic constraint function based on error norm gain, and a stepwise gain coefficient based on error. The specific method includes the following sub-steps: S61. The expression for the distributed observer of the TSSZNN-DC1 discrete model for undirected communication topology is: in, , and Each connected autonomous vehicle is connected to the virtual signal connected autonomous vehicle. , and The observed values, For time step; , for enhanced Laplace matrix of connected autonomous vehicles, in The Laplace matrix of an undirected graph. ,in Let be the adjacency matrix of an undirected graph. When the... Vehicle and the When the vehicle maintains communication, Otherwise, all values ​​are 0. for diagonal matrix, ,in , , ..., ; For the communication link matrix between interconnected autonomous vehicles and virtual signal interconnected autonomous vehicles, when the first When a vehicle-to-vehicle (V2V) connected autonomous vehicle can communicate with a vehicle-to-virtual (V2V) connected autonomous vehicle, ,otherwise ; , , ; For symbolic functions, the expression is: The time-varying bounded decay coefficient based on topological feedback is expressed as follows: ,in , For the constant to be set, This indicates obtaining the smallest eigenvalue of the matrix. Indicates the time of sampling. Dynamic Constraint Controller for TSSZNN-DC1 Discrete Model The expression is: in , , , These are constants to be set. , , The expression for the error-based stepwise gain coefficient is: , , , , , Let be a constant to be set. Represents the L2 norm; The whole can be represented as a strongly bounded dynamic constraint function based on the error norm gain, where the error norm gain is... The expression is: , Let it be a norm, where This indicates that the boundary can be set. , ,in: , An upper bound can be set for acceleration. A lower bound can be set for acceleration; It is a nonlinear manifold, and its expression is: S62. The expression for the distributed observer of the TSSZNN-DC2 discrete model for directed communication topology is: in The time step; where The Laplace matrix of the directed communication topology graph for interconnected autonomous vehicles is given, and there are two topology cases, as follows: This indicates the vehicle following mode. This is represented as a virtual signal-following mode; Represented as a constant to be set; , , ; For symbolic functions, the expression is: The time-varying bounded decay coefficient based on topological feedback is expressed as follows: ,in , For the constant to be set, , , , , This indicates obtaining the smallest eigenvalue of the matrix. Indicates the time at which the sample value was obtained; Dynamic Constraint Controller for TSSZNN-DC2 Discrete Model The expression is: in , , , These are constants to be set. , , The expression for the error-based stepwise gain coefficient is: , , , , , Let be a constant to be set. Represents the L2 norm; The whole can be represented as a strongly bounded dynamic constraint function based on the error norm gain, where the error norm gain is... The expression is: , Let it be a norm, where This indicates that the boundary can be set. , ,in: , An upper bound can be set for acceleration. A lower bound can be set for acceleration; It is a nonlinear manifold, and its expression is: 。 2. The dynamic constraint control method for connected autonomous vehicle platooning based on TSSZNN-DC1 / 2 according to claim 1, characterized in that, The expression for the longitudinal dynamics system equation of the vehicle in step S1 is: in , and They represent the first The location, speed, actual drive / braking torque, and expected drive / braking torque of the connected autonomous vehicles; and They are respectively and The first derivative; They represent the first The mass of a connected autonomous vehicle, the mechanical efficiency of its powertrain, tire radius, combined aerodynamic drag coefficient, gravity, rolling resistance coefficient, and longitudinal dynamic inertial delay; t represents time.

3. The dynamic constraint control method for connected autonomous vehicle platooning based on TSSZNN-DC1 / 2 according to claim 2, characterized in that, The specific method for processing the longitudinal dynamics system equations of the connected autonomous vehicle and adding disturbances using feedback linearization techniques in step S2 includes the following sub-steps: S21. Constructing the expression for the feedback linearization technique: in Representing gravitational acceleration and the first Linearized control inputs for connected autonomous vehicles; S22. Applying the linearized expression to the longitudinal dynamics system equations of the connected autonomous vehicle, we obtain the linearized longitudinal dynamics system equations of the vehicle, whose expression is: in For the first The acceleration of connected autonomous vehicles; for The first derivative; S23. Add a disturbance to the linearized longitudinal dynamics system equations of the vehicle to obtain the disturbed longitudinal dynamics system equations, the expression of which is: in For the first Interference from connected autonomous vehicles; S24, Order The rewritten equations of the longitudinal dynamic system are obtained, and their expressions are as follows: in For the first Interference following new control inputs in connected autonomous vehicles; For the first A dynamic constraint controller to be designed for connected autonomous vehicles.

4. The dynamic constraint control method for connected autonomous vehicle platooning based on TSSZNN-DC1 / 2 according to claim 3, characterized in that, The expression for the longitudinal dynamics system equation of the virtual signal interconnected autonomous vehicle in step S3 is as follows: in , , and These represent the position, speed, acceleration, and control input of the virtual signal interconnected autonomous vehicle, respectively. , and They are respectively , and The first derivative.

5. The dynamic constraint control method for connected automated vehicle platooning based on TSSZNN-DC1 / 2 according to claim 4, characterized in that, The expression for the discretized longitudinal dynamics system equations of the convoy in step S4 is as follows: in , , , and They represent the first Connected autonomous vehicles Always , , , and The sampled values; The discretized virtual longitudinal dynamic system equations in step S4 are expressed as follows: in , , and These respectively represent virtual signal interconnected autonomous vehicles in Always , , and The sampled values.

6. The dynamic constraint control method for connected automated vehicle platooning based on TSSZNN-DC1 / 2 according to claim 5, characterized in that, The expression for the equations of the connected autonomous vehicle platooning dynamic constraint control system in step S5 is as follows: in ; ; ; ; ; , and They are respectively , and The first derivative; Represents the transpose of a matrix; The total number of connected autonomous vehicles; for interference Its expression is: in This means taking the minimum of the two values. This means taking the maximum of the two values. An upper bound can be set for acceleration. A lower bound can be set for acceleration. This represents Gaussian white noise. The expression for the error function of the equations in the dynamic constraint control system for connected autonomous vehicle platooning is as follows: in , and Each connected autonomous vehicle is connected to the virtual signal connected autonomous vehicle. , and Observed values; , This indicates the designed distance between each connected autonomous vehicle and a virtual signal-connected autonomous vehicle. To design the desired spacing between adjacent vehicles; , and These represent the error values ​​for position, speed, and acceleration of each connected autonomous vehicle relative to the virtual signal connected autonomous vehicle.

7. The dynamic constraint control method for connected autonomous vehicle platooning based on TSSZNN-DC1 / 2 according to claim 1, characterized in that, In step S7, the expressions for the position, velocity, and first-order derivative of the connected autonomous vehicles in the equations of the connected autonomous vehicle platooning dynamic constraint control system obtained by solving the TSSZNN-DC1 and TSSZNN-DC2 discrete models are as follows:

8. The dynamic constraint control method for connected autonomous vehicle platooning based on TSSZNN-DC1 / 2 according to claim 7, characterized in that, The specific method for solving the equations of the dynamic constraint control system for connected autonomous vehicle platooning using the TSSZNN-DC1 and TSSZNN-DC2 discrete models is as follows: make in ; express Iterative update value; for The first derivative; Acceleration for autonomous vehicles connected by virtual signals; The speed of autonomous vehicles connected by virtual signals; Will By substituting these values ​​into the TSSZNN-DC1 or TSSZNN-DC2 discrete model, the position, velocity, and acceleration of each connected autonomous vehicle in either the undirected or directed communication topology can be obtained.

Images