Heterogeneous multi-vehicle-queue fuzzy adaptive optimization control system

Abstract

The invention discloses a heterogeneous multi-vehicle-queue fuzzy adaptive optimization control system which comprises a third-order heterogeneous multi-vehicle-queue model, a fuzzy logic system module, a coupling sliding mode surface module, an optimal queue controller module and a judgment-execution network updating law and adaptive law module. A fuzzy logic system is constructed to serve as an execution network and a judgment network, identification of unknown nonlinear dynamic and constraint cost functions is achieved, and an execution-judgment network optimization control design architecture is formed; by constructing a coupling sliding mode surface and a barrier type cost function based on a specified performance function, the queue stability is realized, the distance error is ensured to be converged to a specified performance boundary, and the minimization of energy consumption is also realized. The problem that in an existing control system for the three-order heterogeneous multi-vehicle queue system, the bicycle stability and the queue stability can be ensured, but the energy consumption cannot be minimized is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle platoon control, and in particular to a fuzzy adaptive optimization control system for a heterogeneous multi-vehicle platoon. Background Art

[0002] As an important carrier of intelligent transportation system, automobiles have brought great convenience to the development of society and human life and production. However, with the rapid development of the transportation industry, it faces a series of severe social problems, such as traffic accidents, energy crisis and environmental pollution. As an effective means to solve the severe problems of urban traffic, the intelligent transportation system has attracted widespread attention from scholars at home and abroad. As a key link in the construction of intelligent transportation system, the platoon control of intelligent vehicles has become the research focus of the construction of intelligent transportation system. However, due to factors such as air resistance and bumpy road surface, the vehicle platoon will cause huge energy consumption during driving. Therefore, in order to reduce the energy consumption of multi-vehicle platoons during long-distance driving, and give the expected vehicle spacing of the vehicle platoon under the minimum energy consumption, so as to achieve low energy consumption, green and safe driving, it has attracted widespread attention.

[0003] During the multi-vehicle platooning process, multiple vehicles need to maintain a constant distance between vehicles on a horizontal road. At present, many achievements have been made in the control research of multi-vehicle platooning systems, but the existing technology still has the following problems:

[0004] First, in the existing research on the control system of multi-vehicle platoon systems, most of the vehicles considered are of the same brand and quality. If the power configurations of different types of vehicles are very different, and when driving on steep slopes and bumpy roads, it is easy to cause low power, which will affect the overall driving efficiency of the platoon and lead to excessive energy consumption. Therefore, from a practical point of view, it is difficult to ensure that each vehicle in a multi-vehicle platoon has the same configuration and the same weight and type of cargo.

[0005] Second, in the existing heterogeneous multi-vehicle platoon system control system, most of them do not consider the situation where the control system has unknown nonlinear dynamics and external disturbances. However, in the actual driving process, the vehicle platoon is often affected by factors such as wind resistance and uneven road surface, resulting in the existence of comprehensive characteristics such as high nonlinearity, uncertainty and external disturbance. The existing platoon control system cannot effectively solve the control design problem of the third-order heterogeneous multi-vehicle platoon module with unknown nonlinear dynamics.

[0006] Third, in the existing heterogeneous multi-vehicle platoon system control system, only the single-vehicle stability and platoon stability of the vehicle platoon system can be ensured, but it cannot ensure that the vehicle spacing error converges to the specified performance boundary, and it cannot guarantee the minimization of the energy consumption of the vehicle platoon. Summary of the invention

[0007] The present invention provides a fuzzy adaptive optimization control system for a heterogeneous multi-vehicle platoon to overcome the above technical problems.

[0008] In order to achieve the above object, the technical solution of the present invention is:

[0009] A fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoons, comprising a third-order heterogeneous multi-vehicle platoon module, a fuzzy logic system module, a coupled sliding surface module, an optimal platoon controller module, and a judgment-execution network update law and an adaptive law module;

[0010] The third-order heterogeneous multi-vehicle platoon module is used to perform heterogeneous vehicle platooning according to the kinematic model of the leading vehicle and the kinematic model of the following vehicle, so as to obtain heterogeneous vehicle platooning information and transmit it to the fuzzy logic system module, the coupled sliding surface module, and the judgment-execution network update law and adaptive law module respectively;

[0011] The heterogeneous vehicle formation information includes the position of the leading vehicle, the speed of the leading vehicle, the acceleration of the leading vehicle, the real-time position of the following vehicle, the real-time speed of the following vehicle, and the real-time acceleration of the following vehicle;

[0012] The coupled sliding surface module is used to obtain the quadratic spacing error of the heterogeneous vehicle formation according to the heterogeneous vehicle formation information, and to construct a coupled sliding surface for ensuring the string stability of the vehicle queue according to the quadratic spacing error, and to transmit the coupled sliding surface to the fuzzy logic system module, the optimal queue controller module, and the judgment-execution network update law and adaptive law module respectively;

[0013] The fuzzy logic system module is used to identify the nonlinear dynamic terms in the following vehicle motion model in the heterogeneous vehicle formation and the partial derivative of the optimal cost function constructed according to the coupled sliding mode surface, and send its output to the optimal queue controller module and the judgment-execution network update law and adaptive law module respectively;

[0014] The judgment-execution network update law and adaptive law module is used to construct the parameter update law and parameter adaptive law of the judgment-execution network through the output of the fuzzy logic system module and the coupled sliding surface, and send it to the optimal queue controller module;

[0015] The optimal queue controller module is used to construct an optimal queue controller for heterogeneous vehicle formations according to the parameter update law and parameter adaptation law of the judgment-execution network, the coupled sliding surface and the output of the fuzzy logic system module.

[0016] Furthermore, the expression of the vehicle following vehicle kinematic model is:

[0017]

[0018] Where: M i represents the mass of the i-th vehicle; s i ,v i ,a i They represent the displacement, velocity and acceleration of the i-th vehicle respectively; f i (a i ,v i ) indicates unknown dynamics, and g i represents an intermediate variable and d i represents an unknown external disturbance, represents a composite disturbance and

[0019] represents the air resistance coefficient; represents the cross-sectional area; Indicates the air density;

[0020] Indicates the actual driving force or braking force of the engine; η i Represents the vehicle mechanical efficiency; i represents the road slope coefficient and represents rolling resistance and δ i represents the road slope; g represents the acceleration due to gravity; represents the rolling resistance coefficient; represents gravity, and

[0021] R i Indicates the radius of the vehicle tire; u i Represents the throttle or brake control torque input; τ i Represents the time constant of dynamic characteristics;

[0022] The expression of the kinematic model of the vehicle leader is:

[0023]

[0024] Where: s 0 ,v 0 ,a 0 They represent the displacement, velocity and acceleration of the pilot vehicle respectively.

[0025] Furthermore, the coupled sliding surface module obtains the secondary spacing error of the heterogeneous vehicle formation according to the heterogeneous vehicle formation information, and constructs a coupled sliding surface for ensuring the string stability of the vehicle queue according to the secondary spacing error, which specifically includes the following steps:

[0026] The quadratic spacing error of the heterogeneous vehicle formation is obtained based on the heterogeneous vehicle formation information, and its expression is:

[0027]

[0028] Where: represents the vehicle length; μ represents the safety factor; s i (i=1,2,…,N) represents the displacement of the i-th vehicle; Γ represents the delay caused by acceleration and braking in the vehicle queue; Λ represents the minimum safe distance of the heterogeneous vehicle formation; represents the absolute value of the maximum feasible acceleration of the heterogeneous vehicle formation;

[0029] A continuous function ξ of the error conversion is defined based on the quadratic spacing error to ensure that the quadratic spacing error converges to a specified performance boundary. i (t), which is expressed as

[0030]

[0031] Where: i (t) represents the performance function, And meet ρ i0 ,ρ i∞ ,a i represents the design parameters and ρ i0 >ρ i∞ >0,a i >0; for continuous function ξ i (t) and the initial condition ξ i (0)∈[0,1), we can get i <1, when When |z i (t)|<ρ i (t);

[0032] According to the continuous function ξ i (t) Define the sliding mode variable S i (t), which is expressed as

[0033]

[0034] Where: i is the design parameter;

[0035] According to the sliding mode variable S i (t) Construct a coupled sliding surface to ensure the string stability of the vehicle platoon, which is expressed as

[0036]

[0037] Where: q is a constant, and q>0; S i+1 (t) represents the sliding mode variable of the i+1th vehicle.

[0038] Furthermore, the method for the fuzzy logic system module to identify the nonlinear dynamic terms in the motion model of the following vehicle in the heterogeneous vehicle formation according to the coupled sliding mode surface specifically includes the following steps

[0039] The nonlinear dynamic term f in the following vehicle motion model in heterogeneous vehicle formation is analyzed by fuzzy logic system module. i (v i ,a i ) to perform identification processing to obtain the fuzzy basis function after identification processing and the unknown parameter weights after identification

[0040] And for the nonlinear dynamic term f in the following vehicle motion model in the heterogeneous vehicle formation i (v i ,a i ) is identified as

[0041]

[0042] Where: Denotes the nonlinear dynamic term f i (v i ,a i ) Fuzzy basis functions for identification processing; Denotes the nonlinear dynamic term f i (v i ,a i ) The unknown parameter weights of the identification process; ∈ fi (Z i ) represents the approximation error and satisfies represents a constant;

[0043] The obstacle cost function is defined according to the coupled sliding surface, and its expression is:

[0044]

[0045] Where: represents an intermediate variable, and , i represents a constant, and ζ i >0;u i (Π i ) represents the control law; Indicates u i (Π i )

[0046] The optimal cost function is constructed according to the obstacle-type cost function, and its expression is:

[0047]

[0048] Where: Ψ(Ω i ) represents the admissible control set; Ω i represents a compact set; represents the optimal control law;

[0049] Based on the Bellman optimization criterion, the HJB equation is designed according to the optimal cost function:

[0050]

[0051] Where: Represents the optimal cost function The partial derivative of i represents an intermediate variable, and represents the Hamiltonian function;

[0052] Solution The optimal control law can be obtained , whose expression is

[0053]

[0054] because is unknown, the fuzzy logic system module is used to find the optimal cost function The partial derivative of Perform identification processing to obtain the fuzzy basis function after identification processing

[0055] And for the optimal cost function The partial derivative of The expression for identification processing is

[0056]

[0057] Where: Indicates that the fuzzy logic system module is used to The processing result; Represents the optimal cost function The partial derivative of Fuzzy basis functions for identification processing; represents unknown weight; ∈ i (Π i ) represents the approximation error and satisfies represents a constant; Π i Represents Π i (t) is an abbreviation of

[0058] Furthermore, the parameter update law and parameter adaptation law of the judgment-execution network are constructed;

[0059] The parameter update law of the judgment-execution network is expressed as

[0060]

[0061]

[0062] The expression of the parameter adaptation law of the judgment-execution network is:

[0063]

[0064] Where: express estimates; represents the evaluation network parameters, represents the execution network parameters, and is estimates; Represents a constant; I m represents the identity matrix; γ ci ,γ ai represents the update rate of the judge-execution network and γ ci >0,γ ai >0;γ fi ,σ fi represents the design parameters, and γ fi >0,σ fi >0; represents an intermediate variable, and

[0065] Furthermore, the optimal queue controller of the heterogeneous vehicle formation is constructed, and its expression is:

[0066]

[0067] Where: represents the intermediate variable,

[0068] and represents the design parameters, and ; δ 10 Represents a constant.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] First, by fully considering the impact of comprehensive characteristics such as external disturbances, high nonlinearity and strong uncertainty on the heterogeneous multi-vehicle platoon system, the fuzzy logic system module is used to identify the nonlinear dynamics in the heterogeneous vehicle platoon system, which has good robustness;

[0071] Second, by defining the quadratic spacing error and designing the performance conversion equation, the spacing error is equivalently transformed through the performance conversion equation and the coupled sliding surface is constructed, ensuring that the spacing error of the vehicle queue converges to the specified performance boundary and minimizing the energy consumption of the vehicle queue;

[0072] Third, the parameter update law and parameter adaptation law of the judgment-execution network are constructed through the judgment-execution network update law and adaptation law module, and the optimal queue controller module is combined to realize the design of the fuzzy adaptive optimization queue controller for heterogeneous vehicle formations, thereby effectively minimizing the energy consumption of heterogeneous multi-vehicle queues while ensuring the stability of the queue. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 creative labor.

[0074] Figure 1 It is a schematic diagram of the structure of the fuzzy adaptive optimization control system of a heterogeneous multi-vehicle platoon according to the present invention;

[0075] Figure 2 Graph showing the distance error and performance function of the heterogeneous multi-vehicle platoon system in this embodiment;

[0076] Figure 3 This is the displacement trajectory diagram of the heterogeneous multi-vehicle platoon system in this embodiment;

[0077] Figure 4 This is a speed trajectory diagram of the heterogeneous multi-vehicle platoon system in this embodiment;

[0078] Figure 5 This is the acceleration trajectory diagram of the heterogeneous multi-vehicle platoon system in this embodiment;

[0079] Figure 6 This is the controller trajectory diagram of the heterogeneous multi-vehicle platoon system in this embodiment;

[0080] Figure 7 Evaluate network parameters for heterogeneous multi-vehicle platoon system in this embodiment Trajectory map;

[0081] Figure 8 The network parameters for the heterogeneous multi-vehicle platoon system in this embodiment are: Trajectory map;

[0082] Fig. 9 is the adaptive parameter of the heterogeneous multi-vehicle platoon system in this embodiment Trajectory map. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0084] This embodiment provides a fuzzy adaptive optimization control system for a heterogeneous multi-vehicle platoon. Figure 1 As shown, it includes a three-order heterogeneous multi-vehicle queue module, a fuzzy logic system module, a coupled sliding surface module, an optimal queue controller module, and a judgment-execution network update law and adaptive law module; its connection relationship is that the input end of the fuzzy logic system module is connected to the output end of the three-order heterogeneous multi-vehicle queue module and the output end of the coupled sliding surface, and the output end is connected to the input end of the judgment-execution network update law and adaptive law module and the optimal queue controller; the input end of the coupled sliding surface is connected to the output end of the three-order heterogeneous multi-vehicle queue module, and the output end is connected to the optimal queue controller module, the judgment-execution network update law and adaptive law module and the input end of the fuzzy logic system module; the output end of the judgment-execution network update law and adaptive law module is connected to the input end of the optimal queue controller; the output end of the optimal queue controller is connected to the input end of the three-order heterogeneous multi-vehicle queue module;

[0085] The third-order heterogeneous multi-vehicle platoon module is used to perform heterogeneous vehicle platooning according to the kinematic model of the leading vehicle and the kinematic model of the following vehicle, so as to obtain heterogeneous vehicle platooning information and transmit it to the fuzzy logic system module, the coupled sliding surface module, and the judgment-execution network update law and adaptive law module respectively;

[0086] The heterogeneous vehicle formation information includes the position of the leading vehicle, the speed of the leading vehicle, the acceleration of the leading vehicle, the real-time position of the following vehicle, the real-time speed of the following vehicle, and the real-time acceleration of the following vehicle;

[0087] Specifically, the expression of the vehicle following vehicle kinematic model is:

[0088]

[0089] Where: M i represents the mass of the i-th vehicle; s i ,v i ,a i They represent the displacement, velocity and acceleration of the i-th vehicle respectively; f i (a i ,vi ) indicates unknown dynamics, and g i represents an intermediate variable and d i represents an unknown external disturbance, represents a composite disturbance and

[0090] represents the air resistance coefficient; represents the cross-sectional area; Indicates the air density;

[0091] Indicates the actual driving force or braking force of the engine; η i Represents the vehicle mechanical efficiency; i represents the road slope coefficient and represents rolling resistance and δ i Indicates the road slope; g indicates the acceleration due to gravity; represents the rolling resistance coefficient; represents gravity, and

[0092] R i Indicates the radius of the vehicle tire; u i Represents the throttle or brake control torque input; τ i Represents the time constant of dynamic characteristics;

[0093] The expression of the kinematic model of the vehicle leader is:

[0094]

[0095] Where: s 0 ,v 0 ,a 0 They represent the displacement, velocity and acceleration of the pilot vehicle respectively;

[0096] The coupled sliding surface module is used to obtain the quadratic spacing error of the heterogeneous vehicle formation according to the heterogeneous vehicle formation information, and to construct a coupled sliding surface for ensuring the string stability of the vehicle queue according to the quadratic spacing error, and transmit the coupled sliding surface to the fuzzy logic system module, the optimal queue controller module, and the judgment-execution network update law and adaptive law module respectively, specifically including the following steps:

[0097] The quadratic spacing error of the heterogeneous vehicle formation is obtained based on the heterogeneous vehicle formation information, and its expression is:

[0098]

[0099] Where: represents the vehicle length; μ represents the safety factor; s i (i=1,2,…,N) represents the displacement of the i-th vehicle; Γ represents the delay caused by acceleration and braking in the vehicle queue; Λ represents the minimum safe distance of the heterogeneous vehicle formation; represents the absolute value of the maximum feasible acceleration of the heterogeneous vehicle formation;

[0100] A continuous function ξ of the error conversion is defined based on the quadratic spacing error to ensure that the quadratic spacing error converges to a specified performance boundary. i (t), which is expressed as

[0101]

[0102] Where: i (t) represents the performance function, And meet ρ i0 ,ρ i∞ ,a i represents the design parameters and ρ i0 >ρ i∞ >0,a i >0; for continuous function ξ i (t) and the initial condition ξ i (0)∈[0,1), we can get i <1, when When |z i (t)|<ρ i (t);

[0103] According to the continuous function ξ i (t) Define the sliding mode variable S i (t), which is expressed as

[0104]

[0105] Where: i is the design parameter;

[0106] For the last car, S N+1 =0, Therefore, in order to ensure the string stability of the vehicle queue and solve S i (t) and S i+1 (t), then according to the sliding mode variable S i (t) Construct a coupled sliding surface to ensure the string stability of the vehicle platoon, which is expressed as

[0107]

[0108] Where: q is a constant, and q>0; Si+1 (t) represents the sliding mode variable of the i+1th vehicle.

[0109] In this embodiment, the coupled sliding surface module uses the secondary spacing error defined by the displacement and speed information in the multi-vehicle platoon system, combined with the specified performance function, to perform error conversion. The converted error is input into the sliding surface, and the coupled sliding surface module is constructed for subsequent work;

[0110] The fuzzy logic system module is used to identify the nonlinear dynamic terms in the following vehicle motion model in the heterogeneous vehicle formation according to the coupled sliding mode surface and to identify the partial derivative of the optimal cost function constructed according to the coupled sliding mode surface, and send its output to the optimal queue controller module and the judgment-execution network update law and adaptive law module respectively, specifically including the following steps

[0111] Due to the nonlinear dynamic i (v i ,a i ) is completely unknown, the fuzzy logic system module is used to process f i (v i ,a i ) to obtain the fuzzy basis function after identification processing and the unknown parameter weights after identification

[0112] And for the nonlinear dynamic term f in the following vehicle motion model in the heterogeneous vehicle formation i (v i ,a i ) is identified as

[0113]

[0114] Where: Denotes the nonlinear dynamic term f i (v i ,a i ) Fuzzy basis functions for identification processing; Denotes the nonlinear dynamic term f i (v i ,a i ) The unknown parameter weights of the identification process; ∈ fi (Z i ) represents the approximation error and satisfies represents a constant;

[0115] In addition, in order to achieve the optimal and performance-constrained control objectives of heterogeneous vehicle platooning, an obstacle-type cost function is defined based on the coupled sliding surface, and its expression is:

[0116]

[0117] Where: represents an intermediate variable, and ζ i represents a constant, and ζ i >0;u i (Π i ) represents the control law; Indicates u i (Π i )

[0118] Based on the existing optimal control algorithm, in order to find the ideal optimal control law , then the cost function needs to be minimized; therefore, the optimal cost function is constructed according to the obstacle-type cost function, and its expression is

[0119]

[0120] Where: Ψ(Ω i ) represents the admissible control set; Ω i represents a compact set; represents the optimal control law;

[0121] Based on the Bellman optimization criterion, the HJB equation is designed according to the optimal cost function:

[0122]

[0123] Where: Represents the optimal cost function The partial derivative of i represents an intermediate variable, and represents the Hamiltonian function; where

[0124]

[0125] By solving The optimal control law can be obtained , whose expression is

[0126]

[0127] because is unknown, the fuzzy logic system module is used to find the optimal cost function The partial derivative of Perform identification processing to obtain the fuzzy basis function after identification processing

[0128] And for the optimal cost function The partial derivative of The expression for identification processing is

[0129]

[0130] Where: Indicates that the fuzzy logic system module is used to The processing result; Represents the optimal cost function The partial derivative of Fuzzy basis functions for identification processing; represents unknown weight; ∈ i (Π i ) represents the approximation error and satisfies represents a constant, Π i Represents Π i (t) is an abbreviation of

[0131]

[0132] This embodiment uses the information such as speed and acceleration in the multi-vehicle platoon system through the fuzzy logic system module, uses adaptive parameters and fuzzy basis functions to identify unknown nonlinear dynamics, and uses the output of the sliding surface module to identify the cost function through the fuzzy logic system, and then transmits it to the optimal platoon controller module for subsequent work;

[0133] The judgment-execution network update law and adaptive law module is used to construct the parameter update law and parameter adaptive law of the judgment-execution network according to the output of the fuzzy logic system module and the coupled sliding surface, and send them to the optimal queue controller module;

[0134] Specifically, the parameter update law and parameter adaptation law of the judgment-execution network are constructed;

[0135] The parameter update law of the judgment-execution network is expressed as

[0136]

[0137] The expression of the parameter adaptation law of the judgment-execution network is:

[0138]

[0139] Where: express estimates; represents the evaluation network parameters, represents the execution network parameters, and is estimates; Represents a constant; I m represents the identity matrix; γ ci ,γ airepresents the update rate of the judge-execution network and γ ci >0,γ ai >0;γ fi ,σ fi represents the design parameters, and γ fi >0,σ fi >0; represents an intermediate variable, and

[0140] In this embodiment, the input of the judgment-execution network update law and adaptive law module is the speed information v of the vehicle in the queue system i (t), acceleration information a i (t) and the output of the coupled sliding surface module Π i The output signals are and , whose function is to reflect the dynamic changes of the judgment-execution network parameters and adaptive parameters in the optimal controller, and they input the dynamic changes of parameters into the optimal controller module;

[0141] The optimal queue controller module is used to construct an optimal queue controller for heterogeneous vehicle formations according to the parameter update law and parameter adaptation law of the judgment-execution network, the coupled sliding surface and the output of the fuzzy logic system module;

[0142] Specifically, the optimal queue controller of the heterogeneous vehicle formation is constructed, and its expression is:

[0143]

[0144] Where: represents the intermediate variable,

[0145] and represents the design parameters, and δ 10 Represents a constant.

[0146] In this embodiment, in the heterogeneous multi-vehicle platoon control, the speed v of the following vehicle is i (t) and acceleration a i (t) is output to the fuzzy module; the follower is displaced by s i (t), speed v i (t) and acceleration a i (t) and the pilot car's 0 (t), speed v 0 (t) and acceleration a 0 (t) The information is output to the coupled sliding surface module; the follower displacement s i (t), speed v i (t), acceleration a i (t) and the output Π of the coupled sliding surfacei (t) is input into the judgment-execution network update law and adaptive law module; the output of the judgment-execution network update law and adaptive law module is and the output Π of the coupled sliding surface i (t) is input to the optimal queue controller module; the optimal queue controller module transmits the output signal to the three-order heterogeneous multi-vehicle queue system. The design goal of the present invention is to make the speed and acceleration of the follower vehicle track the speed and acceleration of the pilot vehicle at the given point, and at the same time, the quadratic spacing error can converge to the specified performance boundary without collision. In addition, the designed fuzzy optimization queue controller can minimize energy consumption.

[0147] Compared with the prior art, the present invention has the following beneficial effects:

[0148] First, the present invention fully considers the impact of external disturbances, high nonlinearity and strong uncertainty on the heterogeneous multi-vehicle platoon system, and uses a fuzzy system to identify the nonlinear dynamics in the platoon system. Therefore, compared with the existing technology, the present invention has better robustness.

[0149] Secondly, the present invention defines a quadratic spacing error and designs a performance conversion equation, transforms the spacing error equivalently through the performance conversion equation, and constructs a coupled sliding surface to ensure that the vehicle queue spacing error converges to the specified performance boundary.

[0150] Third, an obstacle-type cost function is proposed, and a fuzzy system is constructed as a judgment-execution network structure to realize the identification of unknown nonlinear dynamic and constraint cost functions. A fuzzy adaptive optimization queue controller is proposed by applying the gradient descent method and optimal control theory, which effectively minimizes energy consumption while ensuring queue stability.

[0151] The simulation results of this embodiment are as follows Figure 2-9 As shown, it shows that all models in the queue are bounded. Figure 2 Indicates the spacing error z i (t)(i=1,2,3,4) and performance boundary ρ i The trajectory of (t) is Figure 2 It can be seen from that the spacing error converges to the specified performance boundary, and the spacing error will not increase along the platoon at any time t, that is, the platoon stability can be ensured; Figure 3 Represents the displacement curves of all vehicles, from Figure 3 It can be seen that the distance between each vehicle and the vehicle in front is greater than the minimum safe distance, that is, there will be no collision between vehicles; Figure 4-5 represents the velocity and acceleration trajectories of all vehicles, and is given by Figure 4-5 It can be seen that bicycle stability can be ensured; Figure 6represents the trajectory of the optimal controller; Figure 7-8 Representation judgment-execution network parameters and trajectories; Fig. 9 Represents adaptive parameters trajectory.

[0152] 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 replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon, characterized in that: It includes a three-order heterogeneous multi-vehicle platoon module, a fuzzy logic system module, a coupled sliding surface module, an optimal platoon controller module, and a judgment-execution network update law and adaptive law module; The third-order heterogeneous multi-vehicle platoon module is used to perform heterogeneous vehicle platooning according to the kinematic model of the leading vehicle and the kinematic model of the following vehicle, so as to obtain heterogeneous vehicle platooning information and transmit it to the fuzzy logic system module, the coupled sliding surface module, and the judgment-execution network update law and adaptive law module respectively; The heterogeneous vehicle formation information includes the position of the leading vehicle, the speed of the leading vehicle, the acceleration of the leading vehicle, the real-time position of the following vehicle, the real-time speed of the following vehicle, and the real-time acceleration of the following vehicle; The coupled sliding surface module is used to obtain the quadratic spacing error of the heterogeneous vehicle formation according to the heterogeneous vehicle formation information, and to construct a coupled sliding surface for ensuring the string stability of the vehicle queue according to the quadratic spacing error, and to transmit the coupled sliding surface to the fuzzy logic system module, the optimal queue controller module, and the judgment-execution network update law and adaptive law module respectively; The fuzzy logic system module is used to identify the nonlinear dynamic terms in the following vehicle motion model in the heterogeneous vehicle formation and the partial derivative of the optimal cost function constructed according to the coupled sliding mode surface, and send its output to the optimal queue controller module and the judgment-execution network update law and adaptive law module respectively; The judgment-execution network update law and adaptive law module is used to construct the parameter update law and parameter adaptive law of the judgment-execution network according to the output of the fuzzy logic system module and the coupled sliding surface, and send them to the optimal queue controller module; The optimal queue controller module is used to construct an optimal queue controller for heterogeneous vehicle formations according to the parameter update law and parameter adaptation law of the judgment-execution network, the coupled sliding surface and the output of the fuzzy logic system module.

2. A fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon according to claim 1, characterized in that: The expression of the vehicle following vehicle kinematic model is: Where: M i represents the mass of the i-th vehicle; s i ,v i ,a i They represent the displacement, velocity and acceleration of the i-th vehicle respectively; f i (a i ,v i ) indicates unknown dynamics, and represents an intermediate variable and d i represents an unknown external disturbance, represents a composite disturbance and represents the air resistance coefficient; represents the cross-sectional area; Indicates the air density; Indicates the actual driving force or braking force of the engine; η i Represents the vehicle mechanical efficiency; i represents the road slope coefficient and represents rolling resistance and δ i represents the road slope; g represents the acceleration due to gravity; represents the rolling resistance coefficient; represents gravity, and R i Indicates the radius of the vehicle tire; u i Represents the throttle or brake control torque input; τ i Represents the time constant of dynamic characteristics; The expression of the kinematic model of the vehicle leader is: Where: s0, v0, a0 represent the displacement, velocity and acceleration of the pilot vehicle respectively.

3. A fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon according to claim 2, characterized in that: The coupled sliding surface module obtains the secondary spacing error of the heterogeneous vehicle formation according to the heterogeneous vehicle formation information, and constructs a coupled sliding surface for ensuring the string stability of the vehicle queue according to the secondary spacing error, specifically including the following steps: The quadratic spacing error of the heterogeneous vehicle formation is obtained based on the heterogeneous vehicle formation information, and its expression is: Where: represents the vehicle length; μ represents the safety factor; s i (i=1,2,…,N) represents the displacement of the i-th vehicle; Γ represents the delay caused by acceleration and braking in the vehicle queue; Λ represents the minimum safe distance of the heterogeneous vehicle formation; represents the absolute value of the maximum feasible acceleration of the heterogeneous vehicle formation; A continuous function ξ of the error conversion is defined based on the quadratic spacing error to ensure that the quadratic spacing error converges to a specified performance boundary. i (t), which is expressed as Where: i (t) represents the performance function, And meet ρ i0 ,ρ i∞ ,a i represents the design parameters and ρ i0 >ρ i∞ >0,a i >0; for continuous function ξ i (t) and the initial condition ξ i (0)∈[0,1), we can get i <1, when When |z i (t)|<ρ i (t); According to the continuous function ξ i (t) Define the sliding mode variable S i (t), which is expressed as Where: i is the design parameter; According to the sliding mode variable S i (t) Construct a coupled sliding surface to ensure the string stability of the vehicle platoon, which is expressed as Where: q is a constant, and q>0; S i+1 (t) represents the sliding mode variable of the i+1th vehicle.

4. The fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon according to claim 2 is characterized in that: The method for identifying the nonlinear dynamic terms in the motion model of the following vehicle in the heterogeneous vehicle formation by the fuzzy logic system module according to the coupled sliding mode surface specifically includes the following steps: The nonlinear dynamic term f in the following vehicle motion model in heterogeneous vehicle formation is analyzed by fuzzy logic system module. i (v i ,a i ) to perform identification processing to obtain the fuzzy basis function after identification processing and the unknown parameter weights after identification And for the nonlinear dynamic term f in the following vehicle motion model in the heterogeneous vehicle formation i (v i ,a i ) is identified as Where: Denotes the nonlinear dynamic term f i (v i ,a i ) Fuzzy basis functions for identification processing; Denotes the nonlinear dynamic term f i (v i ,a i ) The unknown parameter weights of the identification process; ∈ fi (Z i ) represents the approximation error and satisfies represents a constant; The obstacle cost function is defined according to the coupled sliding surface, and its expression is: Where: Q(Π i (t)) represents an intermediate variable, and ζ i represents a constant, and ζ i >0;u i (Π i ) represents the control law; Indicates u i (Π i ) The optimal cost function is constructed according to the obstacle-type cost function, and its expression is: Where: Ψ(Ω i ) represents the admissible control set; Ω i represents a compact set; represents the optimal control law; Based on the Bellman optimization criterion, the HJB equation is designed according to the optimal cost function: Where: Represents the optimal cost function The partial derivative of i represents an intermediate variable, and represents the Hamiltonian function; Solution The optimal control law can be obtained Its expression is because is unknown, the fuzzy logic system module is used to find the optimal cost function The partial derivative of Perform identification processing to obtain the fuzzy basis function after identification processing And for the optimal cost function The partial derivative of The expression for identification processing is Where: Indicates the use of fuzzy logic system module to The processing result; Represents the optimal cost function The partial derivative of Fuzzy basis functions for identification processing; represents unknown weight; ∈ i (Π i ) represents the approximation error and satisfies represents a constant; Π i Represents Π i (t) is an abbreviation of 5. The fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon according to claim 1 is characterized in that: The parameter update law and parameter adaptation law of the constructed judgment-execution network; Its expression is The parameter update law of the judgment-execution network is expressed as The expression of the parameter adaptation law of the judgment-execution network is: Where: express estimates; represents the evaluation network parameters, represents the execution network parameters, and is estimates; Represents a constant; I m represents the identity matrix; γ ci ,γ ai represents the update rate of the judge-execution network and γ ci >0,γ ai >0;γ fi ,σ fi represents the design parameters, and γ fi >0,σ fi >0; represents an intermediate variable, and 6. The fuzzy adaptive optimization control system for heterogeneous multi-vehicle platoon according to claim 1 is characterized in that: The optimal queue controller of the heterogeneous vehicle formation constructed is expressed as Where: represents the intermediate variable, and represents the design parameters, and δ 10 Represents a constant.

Images