Adaptive inclusion control method and system for autonomous underwater vehicle based on GPIO

Abstract

The present invention provides an adaptive inclusion control method and system for an autonomous underwater vehicle based on GPIO, which relates to the field of inclusion control of autonomous underwater vehicles. The method comprises: establishing a dynamic model of a follower autonomous underwater vehicle under external disturbance; observing unmeasurable state information and disturbance state information in the dynamic model of the follower autonomous underwater vehicle based on a generalized proportional integral observer; determining inclusion synchronization error according to the observed information and the topological information of the autonomous underwater vehicle, and establishing an optimal inclusion controller of the follower autonomous underwater vehicle according to the inclusion synchronization error; using the optimal actual controller in the optimal inclusion controller as the control input of the dynamic model of the follower autonomous underwater vehicle to obtain the position and direction vector of the follower autonomous underwater vehicle. The present invention can realize the inclusion control of the autonomous underwater vehicle under external disturbance.

Description

Technical Field

[0001] The present invention relates to the field of autonomous underwater vehicle inclusion control, and in particular to a GPIO-based adaptive inclusion control method and system for an autonomous underwater vehicle. Background Art

[0002] With the development of marine science and technology innovation, autonomous underwater vehicles (AUVs) are widely used in many fields, such as ocean sampling, pipeline detection, environmental monitoring, etc. In the application scenarios of marine engineering, different work tasks require AUVs to use different control technologies to perform, so different control methods need to be studied to meet work requirements. The rapid development of science and technology has made the application of a single AUV insufficient to meet the needs of marine exploration engineering. Therefore, it is inevitable to use multiple AUVs to perform work tasks simultaneously in marine engineering.

[0003] Inclusive control drives the follower AUVs in the system to enter and maintain the convex hull formed by the leaders under the leadership of multiple leader AUVs by designing a distributed control algorithm. When performing a task, the leader AUV detects the surrounding environment and forms a safe area. The remaining follower AUVs enter the safe area and follow the leader to ensure that the individuals in the system avoid dangerous areas and successfully complete the task. Compared with traditional control problems, inclusive control can better highlight the advantages of distributed collaborative control.

[0004] Faced with the harsh marine environment, the AUV system often suffers from interference from the environment, which leads to system performance degradation. In order to enhance the anti-interference performance of the system, many effective estimation methods have been proposed and achieved certain results, such as disturbance observers and extended state observers. However, due to the unpredictable marine environment, the interference caused by severe weather such as tsunamis on the system will change rapidly over time. Traditional observers can often only estimate constant or stage constant disturbances, and cannot effectively estimate polynomial disturbances. Summary of the invention

[0005] In order to address the deficiencies of the prior art, the purpose of the present invention is to provide an adaptive containment control method and system for an autonomous underwater vehicle based on GPIO, so as to enhance the robustness of the autonomous underwater vehicle to harsh environmental disturbances and realize the containment control of the autonomous underwater vehicle under external disturbances.

[0006] To achieve the above object, according to some embodiments, a first aspect of the present invention provides an adaptive inclusion control method for an autonomous underwater vehicle based on GPIO, comprising:

[0007] Establish the dynamic model of the follower autonomous underwater vehicle under external disturbance;

[0008] Based on the generalized proportional integral observer, the unmeasurable state information and disturbance state information in the dynamic model of the follower autonomous underwater vehicle are observed;

[0009] Determine the included synchronization error according to the observation information and the topological information of the autonomous underwater vehicle, and establish the optimal included controller of the follower autonomous underwater vehicle according to the included synchronization error;

[0010] The optimal actual controller in the optimal inclusion controller is used as the control input of the dynamic model of the follower autonomous underwater vehicle, and the position and direction vector of the follower autonomous underwater vehicle are obtained.

[0011] A second aspect of the present invention provides an adaptive containment control system for an autonomous underwater vehicle based on GPIO, comprising:

[0012] A dynamic model building module is configured to build a dynamic model of the follower autonomous underwater vehicle under external disturbance;

[0013] An observation module is configured to observe unmeasurable state information and disturbance state information in a dynamic model of a follower autonomous underwater vehicle based on a generalized proportional integral observer;

[0014] A controller establishment module is configured to determine an inclusion synchronization error based on observation information and topological information of the autonomous underwater vehicle, and establish an optimal inclusion controller of the follower autonomous underwater vehicle based on the inclusion synchronization error;

[0015] The control module is configured to use the optimal actual controller in the optimal inclusion controller as the control input of the dynamics model of the follower autonomous underwater vehicle to obtain the position and direction vector of the follower autonomous underwater vehicle.

[0016] A third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to complete the steps of the above-mentioned GPIO-based autonomous underwater vehicle adaptive control method.

[0017] A fourth aspect of the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-mentioned GPIO-based autonomous underwater vehicle adaptive control method.

[0018] A fifth aspect of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned GPIO-based autonomous underwater vehicle adaptive control method.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention provides an adaptive inclusive control method and system for an autonomous underwater vehicle based on GPIO. Considering that the external force of the autonomous underwater vehicle will change rapidly over time when it encounters a harsh environment, thereby affecting the control performance of the system, a generalized proportional integral observer GPIO is designed for the autonomous underwater vehicle system, which takes into account the information of the high-order derivatives of the interference and has better estimation accuracy for the disturbance that changes rapidly over time, thereby realizing the observation of the system state of the autonomous underwater vehicle and the external interference;

[0021] The present invention proposes an adaptive optimal backstepping inclusion control strategy, establishes an optimal performance index function based on the inclusion synchronization error, and then establishes an optimal inclusion controller. By designing an optimal performance index function containing the inclusion error, the designed optimal inclusion controller can achieve the optimal control of the system while achieving the inclusion control task of the autonomous underwater vehicle;

[0022] The present invention enhances the learning ability of the learning algorithm and uses the gradient descent method to design the adaptive update rate of the judgment neural network and the execution neural network, so that the designed approximate optimal controller can approach the actual optimal controller under the premise of system stability.

[0023] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0025] Figure 1 is a flow chart of the method of the present invention;

[0026] Figure 2 This is the structural diagram of the vertical plane control system of the autonomous underwater vehicle;

[0027] Figure 3 Implementing a topology graph containing control tasks for an autonomous underwater vehicle system;

[0028] Figure 4 Implementing trajectory graphs including control tasks for autonomous underwater vehicle systems;

[0029] Figure 5 is the observation curve of the total disturbance of the autonomous underwater vehicle;

[0030] Figure 6 is the observation curve of the autonomous underwater vehicle F1 state;

[0031] Figure 7 is the observation curve of the autonomous underwater vehicle F2 state;

[0032] Figure 8 To implement and judge the weight convergence curve of neural network in autonomous underwater vehicle system;

[0033] Fig. 9 is the convergence curve of the adaptive law in the autonomous underwater vehicle system. DETAILED DESCRIPTION

[0034] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0035] Embodiment 1

[0036] Embodiment 1 of the present invention, as Figure 1 As shown, an adaptive inclusion control method for an autonomous underwater vehicle based on GPIO is provided, including:

[0037] Step 1: Establish a dynamic model of the follower autonomous underwater vehicle under external disturbance;

[0038] Step 2: Based on the generalized proportional integral observer, the unmeasurable state information and the disturbance state information in the dynamic model of the follower autonomous underwater vehicle are observed;

[0039] Step 3: determining the included synchronization error according to the observation information and the topological information of the autonomous underwater vehicle, and establishing the optimal included controller of the follower autonomous underwater vehicle according to the included synchronization error;

[0040] Step 4: Use the optimal actual controller in the optimal inclusion controller as the control input of the dynamic model of the follower autonomous underwater vehicle to obtain the position and direction vector of the follower autonomous underwater vehicle.

[0041] The control structure of the control system is as follows Figure 2 As shown, the leader autonomous underwater vehicles are marked as L1 and L2, and the follower autonomous underwater vehicles are marked as F1 and F2. The method provided in this embodiment is described in detail by taking the autonomous underwater vehicle system as an example.

[0042] Step 1: Establish the dynamic model of the follower autonomous underwater vehicle under external disturbance. The physical characteristics of the follower autonomous underwater vehicle are modeled, and the dynamic model of the follower autonomous underwater vehicle is established in the body coordinate system and the inertial system. Its generalized velocity vector is defined in the body coordinate system, and its position vector is defined in the inertial system. The dynamic model of the i-th follower autonomous underwater vehicle established is:

[0043]

[0044] Among them, η i =[x i ,y i ,z i ,ψ i ] T To define the position and direction vector of the follower autonomous underwater vehicle in the body coordinate system, x i ,y i ,z i Represents x, y, z axis, ψ i Represents the yaw angle. v i =[u i ,υ i ,ω i ,r i ] T is the generalized velocity vector of the follower AUV in the body coordinate system, where u i ,υ i ,ω i are the velocities of the follower autonomous underwater vehicle in the lateral, longitudinal and vertical directions, r i Represents the yaw angular velocity.

[0045] is the rotation matrix. i is the inertia matrix, C i (v i ) represents the rigid body Coriolis centripetal force matrix, D i (v i ) represents the damping force matrix, g i (η i ) represents the Archimedean restoring force and torque matrix generated by gravity and buoyancy. τ i Represents the control input. τ di Represents the environmental interference force matrix.

[0046] Let x i1 =η i , The dynamic model is converted into a state model to establish a generalized proportional integral observer. The converted state model is:

[0047]

[0048] Among them, x i1 represents the first state of the i-th follower autonomous underwater vehicle, x i2 represents the second state of the i-th follower autonomous underwater vehicle. is the converted system input, is the external disturbance of the system after conversion, f i (·)=J i M-1 i [(C i +D i )J -1 i x i2 +g i ]-J i J -1 i x i2 is a nonlinear function.

[0049] Step 2: Establish a generalized proportional integral observer (GPIO) to observe the unmeasurable state and disturbance information of the follower autonomous underwater vehicle. The established GPIO is:

[0050]

[0051] in, The states x are i1 ,x i2 The estimate of i0 ,...,l i,p+1 is the observer gain. is an external disturbance, which is assumed to be p-order differentiable. k represents a number from 0 to p-2, and is defined as For i0 ,…,ζ i,p-1 The estimate, Yes i The estimated value is approximated by a fuzzy logic system.

[0052] In the framework of the backstepping method, according to the topological information of the AUV system and the estimated value of the GPIO state, the error conversion function is defined as:

[0053]

[0054] Among them, e i1 is the synchronization error, e i2 is the virtual error. N1 represents the number of follower autonomous underwater vehicles, a ij represents the adjacency information of N1 follower autonomous underwater vehicles, a il represents the adjacency information of the N2 leader autonomous underwater vehicles. j represents the distance from the 1st follower autonomous underwater vehicle to the N1th follower autonomous underwater vehicle, and l represents the distance from the N1+1th leader autonomous underwater vehicle to the N1+N2th leader autonomous underwater vehicle. i1 is the output of the follower autonomous underwater vehicle, x j1is the output of the neighboring autonomous underwater vehicle, y ir For the trajectory of virtual leaders, is the optimal virtual control signal for subsequent design. sgn(·) represents the sign function.

[0055] Step 3: Combine backstepping technology and reinforcement learning algorithm to design the optimal inclusion controller. Based on the system dynamic equation of the given autonomous underwater vehicle system's dynamic model, combined with backstepping, the optimal virtual control signal introduced in the inclusion error is used to introduce the virtual control of the current step from the next state variable in the derivative of the Lyapunov function. According to the principle of various offsets, all current and related terms of the aforementioned variables in the Lyapunov function of the current step are kept non-positive, and the optimal virtual controller is designed, and then the optimal actual controller of the system is recursively designed in the last step.

[0056] Step 3.1: Build the optimal virtual controller.

[0057] Establish the optimal performance indicator function for the first step as follows:

[0058]

[0059] Among them, Ω is a compact set containing the origin, Ψ(Ω) is an admissible control set, α i1 It is a virtual controller. is the optimal virtual controller, is the cost function.

[0060] By taking the derivative of both sides of formula (1.5), we can get the Hamilton-Jacobi-Bellman (HJB) equation as follows:

[0061]

[0062] By solving The optimal virtual controller is obtained as follows:

[0063]

[0064] Will This breaks down into:

[0065]

[0066] in, k i1 is a positive design parameter, is a constant term.

[0067] Substituting formula (1.8) into formula (1.7), we have

[0068]

[0069] The optimal virtual controller is approximated by using a reinforcement learning algorithm based on an execution-critic neural network.

[0070]

[0071] in, yes The estimated value of yes Neural Network Approach Unknown dynamics in is the ideal weight of the neural network O ic1 Neural Network Approach Unknown dynamics in is the ideal execution neural network weight O ia1 Estimated value of S i1 is the fuzzy basis function.

[0072] According to (1.6), the designed approximate HJB function is as follows:

[0073]

[0074] Then, the Bellman residual is obtained as follows:

[0075]

[0076] The design is as follows: - The adaptive update rate of the neural network weights is judged to ensure that z i1 (t)→0:

[0077]

[0078] Among them, λ ic1 To judge the design parameters, λ ia1 To implement the design parameters, is the positive parameter of the design and I is the identity matrix.

[0079] In order to design the system controller and ensure the stability of the system, the Lyapunov function L is constructed based on the error conversion function. i1 (x):

[0080]

[0081] Among them, e i1 is the synchronization error. and The expressions are and

[0082] Taking the derivative of (1.15), we can obtain:

[0083]

[0084] Using Young's inequality and equation (1.14), we have:

[0085]

[0086] in, is a constant term in the system.

[0087] Step 3.2: Build the optimal actual controller.

[0088] Construct the optimal performance indicator function in the second step

[0089]

[0090] in, is the actual controller, is the optimal actual controller, is the cost function.

[0091] The corresponding HJB equation is as follows:

[0092]

[0093] Based on the gradient descent method, the adaptive update rate of the weights of the execution-criteria neural network is designed:

[0094]

[0095] Similar to step 3.1, design the following Lyapunov function:

[0096]

[0097] where e i2 is the virtual error, and The expressions are and

[0098] Taking the derivative of equation (1.20), we can get

[0099]

[0100] Applying the Lyapunov function to determine stability, the optimal actual controller and adaptive law are designed as

[0101]

[0102] in, is the adaptation rate, the neural network For approximation Unknown dynamics in represents the estimated value of the ideal weight of the executed neural network, S i2 is the fuzzy basis function, ρ i ,b i is a positive design parameter.

[0103] Using Young's inequality and combining equations (1.19) and (1.21), we can get

[0104]

[0105] Among them, k i2 ,l ip ,b i ,λ ic2 , are all designable positive parameters.

[0106]

[0107] Based on the stability judgment method of Lyapunov function stability theory, the optimal virtual controller, optimal actual controller and adaptive law designed are analyzed to determine the stability of the AUV system:

[0108] Construct the following Lyapunov function:

[0109]

[0110] in,

[0111] By taking the derivative of the constructed Lyapunov function, we can get

[0112]

[0113] It can be seen that all signals in the AUV system are uniformly ultimately bounded and contain errors that converge to a small neighborhood of zero. Ultimately, the stability of the system is proven.

[0114] Step 4: Based on the backstepping method and simplified optimal control, first, a virtual optimal controller is designed in the first step of the backstepping method, and then, the optimal actual controller is recursively designed in the second step of the backstepping method. The Lyapunov stability principle is used to prove that the designed optimal actual controller can ensure the stability of the AUV system. As the control input of the dynamic model of the follower autonomous underwater vehicle, it is input into the dynamic model of the follower autonomous underwater vehicle and outputs the position and direction vector η of the follower autonomous underwater vehicle.i , so that the trajectory of the follower autonomous underwater vehicle remains within the convex hull formed by the trajectory of the leader autonomous underwater vehicle, thereby achieving the system's containment control task.

[0115] In order to prove the feasibility, effectiveness and correctness of this example, the following simulation experiment is carried out. In this simulation experiment, for an autonomous underwater vehicle system with external disturbances, an adaptive optimal controller based on GPIO and reinforcement learning is designed to achieve optimal containment control of the autonomous underwater vehicle system.

[0116] In the optimal controller design process, the system model parameters are set as follows:

[0117] The virtual leader trajectory is: y 1r =[0.5t0.3sin(1.2t+0.2)+1.2-0.3t], y 1r =[0.5t0.3sin(1.2t+0.2)+1.02-0.3t]; the initial value of the follower system state is: x 11 (0) = x 21 (0) = 0.1, x 12 (0) = x 22 (0)=0.

[0118] The system disturbance is: The relevant parameters of the designed GPIO are: p=2,

[0119] Judging Neural Networks and and execute neural network and They all contain 7 neurons, and the center points of the neural network are evenly distributed on [-3,3]. The initial weights of the neural network are:

[0120] The parameters related to the optimal virtual controller and the optimal actual controller are designed as follows: 11 =30,k 12 =18,k 21 =30,k 12 =4; the adaptive law parameters are: ρ1=ρ2=1, b1=1, b2=2.

[0121] The simulation results are as follows Figure 3-Figure 9 As shown, Figure 3 The communication relationship between the follower autonomous underwater vehicle and the leader autonomous underwater vehicle is depicted. Figure 4The trajectory curve of the follower autonomous underwater vehicle is given. It can be seen that according to the designed control scheme, it can converge into the convex hull formed by the leader autonomous underwater vehicle, and the system's containment control task is successfully achieved. Figure 5-Figure 7 The observation performance of GPIO on the external disturbance and state of the autonomous underwater vehicle is demonstrated. It can be seen that the designed GPIO can achieve accurate observation of high-order polynomial disturbances and accurately estimate the unknown state of the system. Figure 8 It reflects the convergence effect of the execution and judgment neural network weights designed based on reinforcement learning, which can converge to a small neighborhood near the origin. Fig. 9 It shows that the designed fuzzy adaptive law can converge to the neighborhood near the origin.

[0122] All signals in the system are uniformly ultimately bounded, and simulation results demonstrate the effectiveness of the proposed optimal control scheme.

[0123] This embodiment uses backstepping and reinforcement learning technology as the design framework, and studies the problem of adaptive optimal inclusion control for an autonomous underwater vehicle system with high-order external disturbances. In addition, this embodiment assumes that the autonomous underwater vehicle system has unmeasurable states and is affected by unknown high-order disturbance information, making the system more general. By designing GPIO, simultaneous observation of the system state and interference information is achieved. This embodiment solves the problem that the HJB equation is difficult to solve directly by utilizing the approximation ability of the reinforcement learning neural network to nonlinear functions and using the reinforcement learning algorithm to approximate the actual optimal controller, and designs an optimal controller that meets the control requirements, and adjusts the weights of the neural network through the adaptive law, so that the designed approximate optimal controller can approximate the actual optimal controller. Finally, it is verified by simulation that the proposed adaptive optimal inclusion control strategy can ensure that all signals are bounded.

[0124] Embodiment 2

[0125] This embodiment provides an adaptive containment control system for an autonomous underwater vehicle based on GPIO, including:

[0126] A dynamic model building module is configured to build a dynamic model of the follower autonomous underwater vehicle under external disturbance;

[0127] An observation module is configured to observe unmeasurable state information and disturbance state information in a dynamic model of a follower autonomous underwater vehicle based on a generalized proportional integral observer;

[0128] A controller establishment module is configured to determine an inclusion synchronization error based on observation information and topological information of the autonomous underwater vehicle, and establish an optimal inclusion controller of the follower autonomous underwater vehicle based on the inclusion synchronization error;

[0129] The control module is configured to use the optimal actual controller in the optimal inclusion controller as the control input of the dynamics model of the follower autonomous underwater vehicle to obtain the position and direction vector of the follower autonomous underwater vehicle.

[0130] It should be noted here that the modules in this embodiment correspond to the steps of the method in Embodiment 1, and the specific implementation process is the same, which will not be repeated here.

[0131] Embodiment 3

[0132] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to complete the steps of the method in the first embodiment.

[0133] Embodiment 4

[0134] This embodiment provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the method in the first embodiment are completed.

[0135] Embodiment 5

[0136] This embodiment provides a computer program product, including a computer program / instruction, characterized in that the computer program / instruction implements the steps of the method in the first embodiment when executed by a processor.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An adaptive control method for an autonomous underwater vehicle based on GPIO, characterized in that: include: Establish the dynamic model of the follower autonomous underwater vehicle under external disturbance; Based on the generalized proportional integral observer, the unmeasurable state information and disturbance state information in the dynamic model of the follower autonomous underwater vehicle are observed; Determine the included synchronization error according to the observation information and the topological information of the autonomous underwater vehicle, and establish the optimal included controller of the follower autonomous underwater vehicle according to the included synchronization error; The optimal actual controller in the optimal inclusion controller is used as the control input of the dynamic model of the follower autonomous underwater vehicle to obtain the position and direction vector of the follower autonomous underwater vehicle; The dynamic model of the i-th follower autonomous underwater vehicle under the external disturbance is: Among them, η i =[x i ,y i ,z i ,ψ i ] T To define the position and direction vector of the autonomous underwater vehicle in the body coordinate system, x i ,y i ,z i Represents x, y, z axis, ψ i represents the yaw angle, v i =[u i ,υ i ,ω i ,r i ] T is the generalized velocity vector of the autonomous underwater vehicle defined in the body coordinate system, where u i ,υ i ,ω i are the velocities of the autonomous underwater vehicle in the lateral, longitudinal and vertical directions, r i represents the yaw angular velocity, J i (ψ i ) is the rotation matrix, M i is the inertia matrix, C i (v i ) represents the rigid body Coriolis centripetal force matrix, D i (v i ) represents the damping force matrix, g i (η i ) represents the Archimedean restoring force and torque matrix generated by gravity and buoyancy, τ i represents the control input, τ di represents the environmental interference force matrix; Let x i1 =η i , The dynamic model is converted into a state model to establish a generalized proportional integral observer; the converted state model is: Among them, x i1 represents the first state of the i-th follower autonomous underwater vehicle, x i2 represents the second state of the i-th follower autonomous underwater vehicle; is the converted system input, is the external disturbance of the system after conversion, f i (·)=J i M -1 i [(C i +D i )J -1 i x i2 +g i ]-J i J -1 i x i2 is a nonlinear function; The method of observing the unmeasurable state information and the disturbance state information in the dynamic model of the follower autonomous underwater vehicle based on the generalized proportional integral observer comprises: converting the dynamic model of the follower autonomous underwater vehicle into a state model, establishing a generalized proportional integral observer according to the state model, and observing the unmeasurable state information and the disturbance state information in the state model based on the established generalized proportional integral observer; The established generalized proportional integral observer is: in, The states x are i1 ,x i2 The estimate of i0 ,…,l i,p+1 is the observer gain, is an external disturbance, which is p-order differentiable, k represents a number from 0 to p-2, and is defined as For i0 ,…,ζ i,p-1 The estimate, Yes i Estimated value approximated by fuzzy logic system; The expression including the synchronization error is: Among them, e i1 is the synchronization error, e i2 is the virtual error, N1 represents the number of follower autonomous underwater vehicles, a ij represents the adjacency information of N1 follower autonomous underwater vehicles, a il represents the adjacency information of N2 leader autonomous underwater vehicles, j represents the distance from the 1st follower autonomous underwater vehicle to the N1th follower autonomous underwater vehicle, l represents the distance from the N1+1th leader autonomous underwater vehicle to the N1+N2th leader autonomous underwater vehicle, and x i1 is the output of the follower autonomous underwater vehicle, x j1 is the output of the neighboring autonomous underwater vehicle, y ir For the trajectory of virtual leaders, is the optimal virtual control signal, sgn(·) represents the sign function; The optimal actual controller of the optimal inclusive controller of the follower autonomous underwater vehicle established based on the inclusive synchronization error is: in, is the adaptation rate, the neural network To approximate the optimal actual controller Unknown dynamics in represents the estimated value of the ideal weights of the neural network, S i2 is the fuzzy basis function, k i2 ,b i ,ρ i is the design parameter, e i2 is the virtual error.

2. An adaptive control system for an autonomous underwater vehicle based on GPIO, characterized in that: include: A dynamic model building module is configured to build a dynamic model of the follower autonomous underwater vehicle under external disturbance; An observation module is configured to observe unmeasurable state information and disturbance state information in a dynamic model of a follower autonomous underwater vehicle based on a generalized proportional integral observer; A controller establishment module is configured to determine an inclusion synchronization error based on observation information and topological information of the autonomous underwater vehicle, and establish an optimal inclusion controller of the follower autonomous underwater vehicle based on the inclusion synchronization error; A control module is configured to use the optimal actual controller in the optimal inclusion controller as a control input of a dynamics model of the follower autonomous underwater vehicle to obtain a position and direction vector of the follower autonomous underwater vehicle; The dynamic model of the i-th follower autonomous underwater vehicle under the external disturbance is: Among them, η i =[x i ,y i ,z i ,ψ i ] T To define the position and direction vector of the autonomous underwater vehicle in the body coordinate system, x i ,y i ,z i Represents x, y, z axis, ψ i represents the yaw angle, v i =[u i ,υ i ,ω i ,r i ] T is the generalized velocity vector of the autonomous underwater vehicle defined in the body coordinate system, where u i ,υ i ,ω i are the velocities of the autonomous underwater vehicle in the lateral, longitudinal and vertical directions, r i represents the yaw angular velocity, J i (ψ i ) is the rotation matrix, M i is the inertia matrix, C i (v i ) represents the rigid body Coriolis centripetal force matrix, D i (v i ) represents the damping force matrix, g i (η i ) represents the Archimedean restoring force and torque matrix generated by gravity and buoyancy, τ i represents the control input, τ di represents the environmental interference force matrix; Let x i1 =η i , The dynamic model is converted into a state model to establish a generalized proportional integral observer; the converted state model is: Among them, x i1 represents the first state of the i-th follower autonomous underwater vehicle, x i2 represents the second state of the i-th follower autonomous underwater vehicle; is the converted system input, is the external disturbance of the system after conversion, f i (·)=J i M -1 i [(C i +D i )J -1 i x i2 +g i ]-J i J -1 i x i2 is a nonlinear function; The method of observing the unmeasurable state information and the disturbance state information in the dynamic model of the follower autonomous underwater vehicle based on the generalized proportional integral observer comprises: converting the dynamic model of the follower autonomous underwater vehicle into a state model, establishing a generalized proportional integral observer according to the state model, and observing the unmeasurable state information and the disturbance state information in the state model based on the established generalized proportional integral observer; The established generalized proportional integral observer is: in, The states x are i1 ,x i2 The estimate of i0 ,…,l i,p+1 is the observer gain, is an external disturbance, which is p-order differentiable, k represents a number from 0 to p-2, and is defined as For i0 ,…,ζ i,p-1 The estimate, Yes i Estimated value approximated by fuzzy logic system; The expression including the synchronization error is: Among them, e i1 is the synchronization error, e i2 is the virtual error, N1 represents the number of follower autonomous underwater vehicles, a ij represents the adjacency information of N1 follower autonomous underwater vehicles, a il represents the adjacency information of N2 leader autonomous underwater vehicles, j represents the distance from the 1st follower autonomous underwater vehicle to the N1th follower autonomous underwater vehicle, l represents the distance from the N1+1th leader autonomous underwater vehicle to the N1+N2th leader autonomous underwater vehicle, and x i1 is the output of the follower autonomous underwater vehicle, x j1 is the output of the neighboring autonomous underwater vehicle, y ir For the trajectory of virtual leaders, is the optimal virtual control signal, sgn(·) represents the sign function; The optimal actual controller of the optimal inclusive controller of the follower autonomous underwater vehicle established based on the inclusive synchronization error is: in, is the adaptation rate, the neural network To approximate the optimal actual controller Unknown dynamics in represents the estimated value of the ideal weights of the neural network, S i2 is the fuzzy basis function, k i2 ,b i ,ρ i is the design parameter, e i2 is the virtual error.

3. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to complete the steps of the method according to claim 1.

4. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the steps of the method according to claim 1.

5. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to claim 1 are implemented.

Images