Optimal preset time control method of hovercraft under privacy protection
Through nonlinear mathematical model and pre-determined time privacy function, combined with Hamilton-Jacobian-Belmann equation and neural network, the optimal virtual control signal is designed, and the hybrid event triggering mechanism is used to solve the path tracking problem of hovercraft under external interference. It realizes predetermined time control and steady-state accuracy under privacy protection.
Patent Information
- Application Number
- CN202411373546.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2025-01-07
AI Technical Summary
Hovercraft are susceptible to external environment interference when sailing on the water, resulting in poor stability and inaccurate path tracking. It is difficult to achieve consistent control of predetermined time by manual manipulation, and there are dangerous situations such as high-speed rotation and side slip.
The state equation is constructed using nonlinear mathematical model, and the privacy functions and functions are introduced for predetermined time. The optimal virtual control signal is designed through the adaptive dynamic programming method of Hamilton-Jacobian-Belmann equation and neural network architecture, and combined with the hybrid event trigger mechanism to construct it. Control input to achieve approximately optimal virtual control at a predetermined time.
The pre-determined time control of the hovercraft is realized under complex external interference, enhancing encryption capabilities and transient steady-state performance, avoiding unnecessary resource consumption, and ensuring steady-state accuracy and stability time of path tracking.
Smart Images
Figure CN119270638A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of hovercraft, in particular to an optimal predetermined time control method for a hovercraft under privacy protection. Background Art
[0002] At present, there have been certain research results on the heading control of hovercraft at home and abroad, but most of them are controlled by manipulating the air rudder of the hovercraft. In addition, the hovercraft was manually controlled in the early days, mainly relying on the driver to observe the changes in the spatial motion parameters of the hovercraft and directly manipulate the actuator. Since the full-lift hovercraft has poor stability when sailing in wind and waves, it is not easy to obtain good control quality by manual control, and it may even cause dangerous navigation conditions such as high-speed rotation and sideslip of the hovercraft, and even more serious accidents such as capsizing.
[0003] When hovercraft is sailing in the water, it is easily disturbed by the external ocean environment such as wind, waves and currents, and it is very easy to slip, which will affect the accuracy of its path tracking. Therefore, how to achieve the predetermined time consistency control of the hovercraft under privacy protection when it is sailing on the water and is disturbed by the external environment has become an urgent problem to be solved. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides an optimal predetermined time control method for a hovercraft under privacy protection, which can enable the hovercraft subjected to complex external environmental interference to have a predetermined time privacy protection control function during the control process.
[0005] In order to achieve the above technical purpose, the adopted technical solution is: an optimal scheduled time control method for a hovercraft under privacy protection, comprising the following steps: Step 1: Establish a nonlinear mathematical model of the hovercraft. Based on the nonlinear mathematical model, construct the state equation under unknown longitudinal velocity and unknown time-varying external disturbance according to the physical characteristics of the hovercraft; Step 2: Use the predetermined time privacy function to equivalently transform the state equation, and introduce the predetermined time function to obtain the error variable s i1 、s i2 ; Step 3: According to the error variable s i1 , ideal optimal virtual control signal and optimal theoretical method to design the optimal cost function of the first subsystem For the optimal cost function The Hamilton-Jacobi-Bellman equation of the first subsystem is constructed by derivation. According to Bellman's optimal principle, the Hamilton-Jacobi-Bellman equation of the first subsystem and the ideal optimal virtual control signal are derived to obtain the ideal optimal virtual control signal. The ideal optimal virtual control signal is approximated by the adaptive dynamic programming method based on the neural network architecture to obtain a new ideal optimal virtual control signal. The new ideal optimal virtual control signal The execution evaluation neural network is used to approximate the optimal virtual control signal α at the predetermined time. i1 After that, the optimal virtual control signal α is designed to approximate the predetermined time i1 The corresponding adaptive parameters and the update law of the execution-evaluation network estimation weights; Step 4: Design a hybrid event trigger mechanism with dynamic scaling, and construct a trigger control input according to the hybrid event trigger mechanism; Step 5: According to the error variable s i2 , ideal optimal virtual controller and optimal theoretical method to design the optimal cost function of the second subsystem For the optimal cost function The Hamilton-Jacobi-Bellman equation of the second subsystem is constructed by derivation. According to Bellman's optimal principle, the Hamilton-Jacobi-Bellman equation of the second subsystem and the ideal optimal virtual controller are derived to obtain the ideal optimal virtual controller. The ideal optimal virtual controller is approximated by the adaptive dynamic programming method based on the neural network architecture to obtain a new ideal optimal virtual controller. For the new ideal optimal virtual controller The execution evaluation neural network is used to approximate the optimal virtual controller α for the scheduled time. i2 After that, the optimal virtual controller α is designed to approximate the scheduled time i2 The corresponding adaptive parameters and the weight update law of the execution-criteria network estimation.
[0006] The present invention also includes step 6, constructing a complete Lyapunov function: And the stability analysis of its time derivative is obtained Where I represents the number of agents, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors; and Both represent parameter estimation errors, m = 1, 2, κi2 is a positive design parameter.
[0007] The predetermined time privacy function of the present invention is: Protected variables are represented by m=1,2,i=1,2,3,……I,I represents the number of agents, where 0≤T d <∞ indicates the expected convergence time, a i Represents a constant.
[0008] The predetermined time function of the present invention where q i ∈Z + Satisfy 2q i >n i +1,n i is the total number of subsystems, 0<T i <∞ and denote the user-defined pre-allocation time and tracking accuracy, respectively.
[0009] The error variable described in the present invention is where s il Represents the error surface l = 1, 2, r i1 represents the filtering error, the virtual control law α i1 As a filter The output signal of the filter is obtained ι i Represents a time scalar.
[0010] The predetermined time approximate optimal virtual control signal α of the present invention i1 for in, Ξ im is the predetermined time privacy function, m = 1, 2, i = 1, 2, 3, ... I, I represents the number of agents, b i = 1 means that agent i can receive the leader signal, otherwise, b i =0,a ij = 1 means that the i-th agent can receive the information of the neighboring agent j, otherwise a ij =0;θ i1 is a positive design parameter; represents the basis function of the neural network approximating the first unknown function, Represents a positive constant, for variable s i1 After taking the derivative, we get σ i1 ; for The estimated value of represents the weight vector of the neural network, Represents the approximation error of the neural network; express The estimated value of is the optimal weight vector of the neural network The corresponding execution optimal weight vector.
[0011] The predetermined time approximate optimal virtual control signal α of the present invention i1 Corresponding adaptive parameters and weight update law of execution-evaluation network estimation Among them, l i1 ,γ ic1 ,γ ia1 For positive design parameters, γ ic1 >γ ia1 / 2, for variable s i1 After taking the derivative, we get σ i1 , represents a normal quantity, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors, represents the weight vector of the neural network, represents the approximation error of the neural network, b i = 1 means that agent i can receive the leader signal, otherwise, b i =0,a ij = 1 means that the i-th agent can receive the information of the neighboring agent j, otherwise a ij =0; is the basis function for the neural network to approximate the first unknown cost function.
[0012] The specific implementation of step 4 of the present invention is a hybrid event triggering mechanism: Where t represents the system running time, Indicates the last triggering moment. is the measurement error signal, Represents the trigger control input, t i(k+1) Indicates the next triggering time, t i(k+1) >t ik ,k∈N + , 0<ν i <1,Λ i ,κ i ,D i1 ,D i2 ,and All are normal numbers, 0<D i3 (t)<D i3 (0), D i3 (0) is D i3 (t) Initial value state, o i >0 means the preset accuracy determines when D i3 (t) is restarted to its original value, ν i |u i (t)|+Λ i Indicates the control signal gap between two consecutive trigger constants. and ∈ i Indicates the rate of change of the control signal gap, z i1 represents the consistency error; Through the mixed event trigger mechanism, we can get (1) When When the trigger control input is constructed Among them, i >0 and (2) When When the trigger control input is constructed in, u i (t) is the control input.
[0013] The predetermined time approximate optimal virtual controller α of the present invention i2 for in, Ξ im is the predetermined time privacy function, m = 1, 2, i = 1, 2, 3, ... I, I represents the number of agents, Respectively The estimated value of is the optimal weight vector of the neural network The corresponding execution optimal weight vector, Represents the basis function of the neural network approximating the second unknown function, for the variable s i1 After taking the derivative, we get σ i2 ,θ i2 ,κ i2 is a positive design parameter, τ i ,o i ,E i4 Indicates normal amount, ω i2 g i2 The upper bound value, g i2 =(o i2 / k i ), o i2 represents the fluid dynamic damping coefficient, k i Represents the moment of inertia.
[0014] The predetermined time approximate optimal virtual controller α of the present invention i2 Corresponding adaptive parameters and weight update law of execution-evaluation network estimation Among them l i2 ,γ ic2 ,γ ia2 For positive design parameters, γ ic2 >γ ia2 / 2, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors, represents the basis function of the neural network approximating the second unknown function, For normal amount, is the basis function for the neural network to approximate the second unknown cost function.
[0015] The beneficial effect of the present invention is that: when confidential state information needs to be protected within a specified time, the scheme introduces a predetermined time privacy function and a predetermined time function to enhance encryption capability and transient and steady-state performance. Technically, an intermediate variable Q is constructed. i1To solve the problem of unknown longitudinal velocity, the unknown time-varying external disturbance is treated as an unknown function, so that the scheduled time optimization of the controller is achieved in the backstepping control process based on reinforcement learning. Subsequently, a novel Lyapunov energy function is used to eliminate the disadvantage of semi-global boundedness of the error surface caused by dynamic surface control. It is worth noting that by utilizing a unique hybrid event triggering mechanism, unnecessary resource consumption and monitoring conditions at any time are effectively avoided. This scheme solves the path tracking problem caused by external environmental interference when the hovercraft is sailing on the water surface at an unknown longitudinal velocity, and realizes the scheduled time control of the hovercraft under privacy protection, which includes the specified steady-state accuracy and stabilization time. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 It is a schematic diagram of the process of the hovercraft of the present invention.
[0017] Figure 2 It is the communication topology diagram of the hovercraft of the present invention.
[0018] Figure 3 This is a diagram of the simulation results of the lateral velocity trajectory tracking of the hovercraft of the present invention.
[0019] Figure 4 This is a simulation result diagram of the protected consistency error of the hovercraft of the present invention.
[0020] Figure 5 It is the simulation result of the hovercraft triggering interval. DETAILED DESCRIPTION
[0021] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0022] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.
[0023] At the same time, it should be noted that: unless otherwise defined, the technical terms or scientific terms used in this document should have the common meanings understood by persons with ordinary skills in the technical field to which the present invention belongs.
[0024] like Figure 1 , Figure 2 As shown, an optimal scheduled time control method for a hovercraft under privacy protection includes multiple steps. The technical solutions of each step are described in detail below: Step 1: First, mathematically model the hovercraft, and then construct the unknown control coefficients and state equations under disturbance according to the physical characteristics of the hovercraft.
[0025] The specific steps include: Step 1.1, first establish the nonlinear mathematical model of the hovercraft: Among them, ξ i and p i denote the lateral speed and the rotational speed respectively, Yes i The time derivative of Yes i The time derivative of . i1 ,o i2 and i3 represents the fluid dynamic damping coefficient. k i Represents the moment of inertia. ix and τ id They represent the unknown longitudinal velocity and the unknown time-varying external disturbance respectively. i represents the torque caused by the air rudder. I represents the total number of agents, and i=1,2,3,4 are selected in the experiment.
[0026] Step 1.2, define variable x i1 =ξ i ,x i2 =o i2 p i ,g i1 =Ψ ix and g i2 =(o i2 / k i ), the motion model of the hovercraft can be expressed as the following state space equation: Among them i1 o i2 o i3 It can be set according to the state of the hovercraft. In the experiment, the hovercraft parameter is selected as o i1 =1,o i2 =2 and o i3 =k i =20.
[0027] Step 2: Use the predetermined time privacy function to equivalently transform the state equation of the hovercraft, and introduce the predetermined time function to obtain the error variable.
[0028] The specific steps include: Step 2.1: Privacy function based on scheduled time Protected variables are represented by m=1,2,where 0≤T d <∞ indicates the expected convergence time, a i Represents a constant.
[0029] Step 2.2: In the preparation process of optimal control, the state equation of the initial system can be converted into: in represents the unknown external disturbance, and Indicates the variables contained in the function. Indicates Ξ im The derivative of yes The time derivative of .
[0030] Step 2.3: Introduce the scheduled time function where q i ∈Z + Satisfy 2q i >n i +1,n i Represents the number of subsystems, where n i =2.0<T i <∞ and denote the user-defined pre-allocation time and tracking accuracy, respectively.
[0031] In order to achieve the specified time consistency tracking control, the protected consistency error meets the following conditions: where z i1 represents the consistency error variable, v i A set of nodes representing a communication topology network. ij = 1 means that the i-th agent can receive information from the adjacent j-th agent, otherwise a ij = 0. b i =1 means the i-th agent can accept the leader signal, otherwise, b i =0. Represents the protection signal received by the leader, represents the protection signal received from neighboring agent j.
[0032] In addition, the time-varying function is defined as: in and c i>0 indicates positive scale and pre-assigned safe distance, q i is a designed positive integer, for example: q i =2.
[0033] Defining Transformation Variables
[0034] Step 2.4, define the error coordinates to obtain the error variable: where s il (l=1,2) represents the error surface, r i1 represents the filtering error, the virtual control law α i1 As a filter The output signal of the filter is obtained ι i Represents the time scalar. Step 3: According to the error variable s i1 , ideal optimal virtual control signal and optimal theoretical method, construct the Hamilton-Jacobi-Bellman equation of the first subsystem, and solve the approximate optimal virtual control signal α at the predetermined time i1 And the corresponding adaptive parameters and the execution-evaluation network estimated weight update law
[0035] The specific steps include: Step 3.1: Based on the error variable s i1 , ideal optimal virtual control signal and optimal theoretical method, design the optimal cost function: Where Ω x represents the set of inputs for admissible control, Minimize The ideal optimal virtual control signal, s i1 (τ) is abbreviated as s i1 , Abbreviated as
[0036] Step 3.2, derive the following Hamilton-Jacobi-Bellman equation: in a ij = 1 means that the i-th agent can receive information from the adjacent j-th agent, otherwise a ij = 0. b i = 1 means that agent i can receive the leader signal, otherwise, b i =0. in, for The time derivative of for The time derivative of σ i1 σ i2 is the variable φ i After derivation, we get . Where Ω i (Γ i (t)) is abbreviated as Ω i .
[0037] Step 3.3: According to Bellman's optimal principle, by solving The ideal optimal virtual control signal is: The intermediate variable is defined as
[0038] The ideal optimal virtual control signal of the above formula is approximately solved by an adaptive dynamic programming method based on a radial basis function neural network architecture to obtain a new ideal optimal controller for controlling the air cushion craft. in is an auxiliary function, represents the basis function of the neural network approximating the first unknown function, represents the weight vector of the neural network, represents the approximation error of the neural network, represents a positive constant, θ i1 is a positive design parameter.
[0039] Step 3.4: The optimal gradient function The decomposition can be obtained as follows: Step 3.5: Use radial basis neural network to solve the unknown cost function Approximation processing, in is the ideal weight vector, is the basis function for the neural network to approximate the first unknown cost function, is the approximation error, satisfying the relation is an arbitrarily small positive constant, and a reinforcement learning strategy with an execution-evaluation structure is adopted to execute the evaluation neural network to approximate the optimal gradient function and the optimal virtual control signal α at a predetermined time.i1 : in and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors.
[0040] Step 3.6: Design adaptive parameters and the execution-evaluation network estimated weight update law for: Among them l i1 ,γ ic1 ,γ ia1 For positive design parameters, γ ic1 >γ ia1 / 2.
[0041] Step 4: Design a hybrid event trigger mechanism with dynamic scaling, and construct a trigger control input according to the hybrid event trigger mechanism to control the hovercraft.
[0042] The specific steps include: Step 4.1: In order to reduce the amount of data transmission and avoid continuous monitoring of the trigger criteria, a hybrid event trigger mechanism with dynamic scaling is designed: Where t represents the system running time, Indicates the last triggering moment. is the measurement error signal, Represents the trigger control input, t i(k+1) Indicates the next triggering time. In addition, t i(k+1) >t ik ,k∈N + , 0<ν i <1,Λ i ,κ i ,D i1 ,D i2 ,and are all positive numbers. 0<D i3(t)<D i3 (0), D i3 (0) is D i3 (t) Initial value state. i >0 represents the preset precision, which determines when D i3 (t) is reset to its original value. In addition, ν i |u i (t)|+Λ i Indicates the control signal gap between two consecutive trigger constants. and ∈ i The overall represents the rate of change of the control signal gap. According to the consistency error z i1 The size of the trigger signal is updated.
[0043] By analyzing the trigger mechanism of mixed events, we can get (1) When When the trigger control input is constructed where ρ i >0 and Assume that there is a continuous time-varying scalar Satisfaction relationship get Therefore, we can finally get s i2 u i ≤-|s i2 α i2 |+0.557ρ i .
[0044] (2) When When the trigger control input is constructed Assume that there exists a scalar E i4 satisfy exist Available
[0045] Therefore, we can finally get in
[0046] Step 5: Similar to the optimization process in step 3, according to the error variable s i2 , ideal optimal virtual controller and optimal theoretical method, construct the Hamilton-Jacobi-Bellman equation of the second subsystem, and solve the approximate optimal virtual controller α for the predetermined time i2 And the corresponding adaptive parameters and the execution-evaluation network estimated weight update law
[0047] Step 5.1. Design the optimal virtual controller for the second subsystem with a predetermined time approximation: where θ i2 ,κ i2 is a positive design parameter, τ i ,o i ,E i4 Indicates normal amount, and Respectively and The estimated value of ω i2 g i2 The upper bound value of represents the basis function of the neural network approximating the second unknown function, is the basis function for the neural network to approximate the second unknown cost function, and the neural network input is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors;
[0048] Step 5.2, design the adaptive law and execute-evaluate the network to estimate the weight update law: Among them l i2 ,γ ic2 ,γ ia2 For positive design parameters, γ ic2 >γ ia2 / 2.
[0049] Step 6: Construct a complete Lyapunov function: in and Both represent parameter estimation errors, and stability analysis of their time derivatives yields Prove that the hovercraft system achieves closed-loop stability, To define a scalar, Is a normal number.
[0050] According to Lyapunov's stability theorem, optimal control of the hovercraft can be achieved by adjusting parameters. The closed-loop system is a traditional concept in the field of control, that is, a feedback control system. The present invention proves that the hovercraft achieves closed-loop stability, solves the path tracking problem caused by the interference of the external environment when the hovercraft is sailing on the water surface at an unknown longitudinal speed, and realizes the predetermined time control of the hovercraft under privacy protection and the specified steady-state accuracy and stable time control.
[0051] The simulation was carried out by Matlab, and its stability and real-time performance were analyzed. The control purpose of the simulation experiment is to make the lateral velocity trajectory of the four hovercraft track the reference trajectory of the leader, and to achieve the specified steady-state accuracy and stable time control.
[0052] The initial conditions of the simulation are: Design parameters: Gaussian function: i = 1, 2, 3, 4, m = 1, 2, ι im =0,±1,±2,±3,±4, The reference trajectory is
[0053] The design goal of this application is to design an optimal scheduled time control method for hovercraft under privacy protection. The invention solves the path tracking problem caused by the interference of the external environment when the hovercraft sails on the water surface with unknown longitudinal speed, and realizes the scheduled time control and specified steady-state accuracy and stable time control of the hovercraft under privacy protection. Figure 3 shown. Figure 4 Simulation results of protected consistency error for hovercraft. Figure 5 Simulation results for the hovercraft trigger interval.
[0054] It should be noted that the parts not described in detail in the above embodiments are all prior art.
[0055] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention should be covered within the protection scope of the present invention.
Claims
1. An optimal scheduled time control method for a hovercraft under privacy protection, characterized in that: The following steps are involved: Step 1: Establish a nonlinear mathematical model of the hovercraft. Based on the nonlinear mathematical model, construct the state equation under unknown longitudinal velocity and unknown time-varying external disturbance according to the physical characteristics of the hovercraft; Step 2: Use the predetermined time privacy function to equivalently transform the state equation, and introduce the predetermined time function to obtain the error variable s i1 、s i2 ; Step 3: According to the error variable s i1 , ideal optimal virtual control signal and optimal theoretical method to design the optimal cost function of the first subsystem For the optimal cost function The Hamilton-Jacobi-Bellman equation of the first subsystem is constructed by derivation. According to Bellman's optimal principle, the Hamilton-Jacobi-Bellman equation of the first subsystem and the ideal optimal virtual control signal are derived to obtain the ideal optimal virtual control signal. The ideal optimal virtual control signal is approximated by the adaptive dynamic programming method based on the neural network architecture to obtain a new ideal optimal virtual control signal. The new ideal optimal virtual control signal The execution evaluation neural network is used to approximate the optimal virtual control signal α at the predetermined time. i1 After that, the optimal virtual control signal α is designed to approximate the predetermined time i1 The corresponding adaptive parameters and the update law of the execution-evaluation network estimation weights; Step 4: Design a hybrid event trigger mechanism with dynamic scaling, and construct a trigger control input according to the hybrid event trigger mechanism; Step 5: According to the error variable s i2 , ideal optimal virtual controller and optimal theoretical method to design the optimal cost function of the second subsystem For the optimal cost function The Hamilton-Jacobi-Bellman equation of the second subsystem is constructed by derivation. According to Bellman's optimal principle, the Hamilton-Jacobi-Bellman equation of the second subsystem and the ideal optimal virtual controller are derived to obtain the ideal optimal virtual controller. The ideal optimal virtual controller is approximated by the adaptive dynamic programming method based on the neural network architecture to obtain a new ideal optimal virtual controller. For the new ideal optimal virtual controller The execution evaluation neural network is used to approximate the optimal virtual controller α for the scheduled time. i2 After that, the optimal virtual controller α is designed to approximate the scheduled time i2 The corresponding adaptive parameters and the weight update law of the execution-criteria network estimation.
2. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: It also includes step 6, constructing a complete Lyapunov function: And the stability analysis of its time derivative is obtained Where I represents the number of agents, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors; and Both represent parameter estimation errors, m = 1, 2, κ i2 is a positive design parameter.
3. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time privacy function is Protected variables are represented by m=1,2,i=1,2,3,……I,I represents the number of agents, where 0≤T d <∞ indicates the expected convergence time, a i Represents a constant.
4. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time function where q i ∈Z + Satisfy 2q i >n i +1,n i is the total number of subsystems, 0<T i <∞ and denote the user-defined pre-allocation time and tracking accuracy, respectively.
5. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The error variable is where s il Represents the error surface l = 1, 2, r i1 represents the filtering error, the virtual control law α i1 As a filter The output signal of the filter is obtained ι i Represents a time scalar.
6. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time approximates the optimal virtual control signal α i1 for in, Ξ im is the predetermined time privacy function, m = 1, 2, i = 1, 2, 3, ... I, I represents the number of agents, b i = 1 means that agent i can receive the leader signal, otherwise, b i =0,a ij = 1 means that the i-th agent can receive the information of the neighboring agent j, otherwise a ij =0;θ i1 is a positive design parameter; represents the basis function of the neural network approximating the first unknown function, represents a positive constant, for variable s i1 After taking the derivative, we get σ i1 ; for The estimated value of represents the weight vector of the neural network, Represents the approximation error of the neural network; express The estimated value of is the optimal weight vector of the neural network The corresponding execution optimal weight vector.
7. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time approximates the optimal virtual control signal α i1 Corresponding adaptive parameters and weight update law of execution-evaluation network estimation Among them, l i1 ,γ ic1 ,γ ia1 For positive design parameters, γ ic1 >γ ia1 / 2, for variable s i1 After taking the derivative, we get σ i1 , represents a normal quantity, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors, represents the weight vector of the neural network, represents the approximation error of the neural network, b i = 1 means that agent i can receive the leader signal, otherwise, b i =0,a ij = 1 means that the i-th agent can receive the information of the neighboring agent j, otherwise a ij =0; is the basis function for the neural network to approximate the first unknown cost function.
8. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The specific implementation of step 4 is a hybrid event trigger mechanism: Where t represents the system running time, Indicates the last triggering moment. is the measurement error signal, Represents the trigger control input, t i(k+1) Indicates the next triggering time, t i(k+1) >t ik ,k∈N + , 0<ν i <1,Λ i ,κ i ,D i1 ,D i2 ,and All are normal numbers, 0<D i3 (t)<D i3 (0), D i3 (0) is D i3 (t) Initial value state, o i >0 means the preset accuracy determines when D i3 (t) is restarted to its original value, ν i |u i (t)|+Λ i Indicates the control signal gap between two consecutive trigger constants. and ∈ i Indicates the rate of change of the control signal gap, z i1 represents the consistency error; Through the mixed event trigger mechanism, we can get (1) When When the trigger control input is constructed Among them, i >0 and (2) When When the trigger control input is constructed in, u i (t) is the control input.
9. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time approximates the optimal virtual controller α i2 for in, Ξ im is the predetermined time privacy function, m = 1, 2, i = 1, 2, 3, ... I, I represents the number of agents, Respectively The estimated value of is the optimal weight vector of the neural network The corresponding execution optimal weight vector, Represents the basis function of the neural network approximating the second unknown function, for the variable s i1 After taking the derivative, we get σ i2 ,θ i2 ,κ i2 is a positive design parameter, τ i ,o i ,E i4 Indicates normal amount, g i2 The upper bound value, g i2 =(o i2 / k i ), o i2 represents the fluid dynamic damping coefficient, k i Represents the moment of inertia.
10. The optimal scheduled time control method for a hovercraft under privacy protection as claimed in claim 1, characterized in that: The predetermined time approximates the optimal virtual control signal α i2 Corresponding adaptive parameters and weight update law of execution-evaluation network estimation Among them l i2 ,γ ic2 ,γ ia2 For positive design parameters, γ ic2 >γ ia2 / 2, and Respectively and The estimated value of is the optimal weight vector of the neural network The corresponding execution and evaluation optimal weight vectors, represents the basis function of the neural network approximating the second unknown function, For normal amount, is the basis function for the neural network to approximate the second unknown cost function.