A distributed finite-time data-driven heading consistency control method for unmanned ships

Abstract

The present invention discloses a distributed finite-time unmanned ship data-driven heading consistency control method, comprising: constructing a dynamic model of each follower of the unmanned ship system; selecting a finite-time preset performance function and obtaining preset performance constraint parameters; constructing a data-driven fuzzy predictor according to a fuzzy logic system, the dynamic model of each follower of the unmanned ship system and data stored in a data stack; constructing a second-order nonlinear tracking differentiator; constructing a distributed finite-time consistency controller according to the dynamic model, the preset performance function, the preset performance constraint parameters, the data-driven fuzzy predictor and the second-order nonlinear tracking differentiator, and combining the fuzzy logic system, the filter and other parameters to obtain a distributed finite-time unmanned ship data-driven heading consistency controller to realize the unmanned ship heading consistency control; the present invention can effectively suppress the overshoot of the synchronization error, improve the transient and steady-state performance of the system, achieve rapid response and control effect and is more in line with practical applications.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-agent collaborative control, and in particular to a distributed finite-time unmanned ship data-driven heading consistency control method. Background Art

[0002] The ocean's depths hold abundant untapped resources, the rational utilization of which is of great significance for promoting the sustainable development of human society. The ocean is not only a critical route for global trade but also a frontier for safeguarding national security and strategic interests. The development of the marine economy will help us better address global challenges such as population growth, resource constraints, and environmental degradation, and its strategic value is self-evident. Unmanned vessels, with their flexible maneuverability, exceptional adaptability, advanced automation, and low manufacturing costs, are becoming a revolutionary force in marine transportation. Capable of remote control or autonomous navigation, they offer unprecedented opportunities for the exploration and development of marine resources. However, with the increasing complexity of the marine environment and the diversity of mission requirements, the limitations of a single unmanned vessel are becoming apparent. In this context, collaborative unmanned vessel operations demonstrate their unique advantages, enabling them to more efficiently and flexibly address various challenges.

[0003] In the field of unmanned ship control, the problem of cooperative control of unmanned ships has become a hot topic and focus of research in recent years. Among them, consistency control, as the core of cooperative control, has attracted the attention of many scholars. Unmanned ship cluster consistency control not only improves the efficiency and reliability of task execution, but also enhances the adaptability to complex marine environments, reduces labor costs, improves operational efficiency, and reduces the risk of casualties when performing high-risk tasks, providing strong technical support for the sustainable development of marine resources and the protection of the marine environment. However, although scholars have proposed a variety of feasible control methods, the existing unmanned ship consistency control technology still has some shortcomings:

[0004] First, existing unmanned ship heading consistency control methods rarely consider the problem of finite-time convergence. When practical applications require high transient and steady-state performance of the system or require fast response, failure to consider this problem may cause the overshoot of the synchronization error to be too large, affecting the control effect of the system.

[0005] Second: Distributed clusters are often used in the existing unmanned ship heading consistency adaptive control. However, when computing resources are limited, communication range is short, communication bandwidth is narrow, and the intelligent agents to be manipulated and controlled are large, distributed clusters are difficult to meet control requirements.

[0006] Third: Existing finite-time consistency control methods rarely consider the issue of relaxing continuous excitation conditions. However, in practical applications, due to various restrictions, the system's continuous excitation signal is often difficult to monitor, so there is an urgent need to develop a feasible method to solve this problem. Summary of the Invention

[0007] The present invention provides a distributed finite-time unmanned ship data-driven heading consistency control method to overcome the problems in existing unmanned ship heading consistency control methods, such as the failure to consider finite-time convergence and the relaxation of continuous excitation conditions, and the use of distributed clusters, which leads to the inability to adapt to higher and more comprehensive system requirements in practical applications and the technical problems of limited control effect.

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

[0009] A distributed finite-time data-driven heading consistency control method for an unmanned ship, comprising:

[0010] S1: Introducing a dynamic model of the unmanned ship system's heading characteristics, and constructing a dynamic model of each follower of the unmanned ship system based on the dynamic model of the unmanned ship system's heading characteristics, to obtain an unmanned ship heading angle state signal and an unmanned ship heading angle state derivative signal;

[0011] S2: Select a finite time preset performance function and obtain preset performance constraint parameters;

[0012] S3: introducing a fuzzy logic system, a filter and a data stack, filtering the unmanned ship heading angle state derivative signal and the fuzzy basis vector in the fuzzy logic system, obtaining filtered data and saving it through the data stack;

[0013] S4: constructing a data-driven fuzzy predictor based on the fuzzy logic system, the dynamic model of each follower of the unmanned ship system, and the data stored in the data stack, wherein the data-driven fuzzy predictor is used to receive the estimated value of the fuzzy weight sent by the fuzzy logic system and predict the approximate value of the unknown nonlinear function;

[0014] S5: introducing a virtual control law based on a dynamic surface, and constructing a second-order nonlinear tracking differentiator according to the virtual control law based on the dynamic surface; the second-order nonlinear tracking differentiator is used to receive the approximation value of the unknown nonlinear function and the virtual control law from the data-driven fuzzy predictor;

[0015] S6: constructing a distributed finite-time consistency controller according to the dynamic model of each follower of the unmanned ship, a preset performance function, preset performance constraint parameters, a data-driven fuzzy predictor, and a second-order nonlinear tracking differentiator to perform finite-time consistency control on the unmanned ship;

[0016] S7: Combining the fuzzy logic system, the second-order nonlinear tracking differentiator, the data stack, the filter, the data-driven fuzzy predictor, the unmanned ship system and the distributed finite-time consistency controller, a distributed finite-time unmanned ship data-driven heading consistency controller is obtained to achieve the unmanned ship heading consistency control.

[0017] Furthermore, S6 builds a distributed finite-time consistency controller, including:

[0018] The distributed finite-time consistency controller is shown in formulas (1), (2) and (3),

[0019]

[0020] Where: α i,1 is the virtual control law based on the dynamic surface, d i represents the sum of the degrees between the follower and its neighbors, represents the degree of the jth neighbor leader associated with the i-th follower drone ship; a i,j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i,j =1; otherwise a i,j =0;a i,r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i,r =1; otherwise a i,r =0;Δ i,1 , Δ i,2 represents the control gain of the data-driven fuzzy predictor; o i,1 represents the design parameters, C i,0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, χ i,1 Preset performance function for finite time, Represents the preset performance function χ i,1 The derivative of n i,1 is the preset performance constraint parameter; x j,2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; y r represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i,1 , W i,2 represents the adaptive parameter, It's W i,1 The estimated value of It's W i,2 The estimated value of η i,1 , η i,2 represents the blurred basis vector after filtering; n i,2 represents the dynamic surface error, x i,2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship, It is x i,2 The estimated value of represents the estimation error; represents the filtered virtual control law based on the dynamic surface; is the adaptive gain; is the adaptive parameter; It means t=t p Time μ i,2 and v i,2 The value of Indicates the stack length; represents the filtered second-order nonlinear tracking differentiator; for The transpose of W i,2 The transpose of u i Represents the control input of the i-th follower unmanned ship system.

[0021] Furthermore, S1 introduces a dynamic model of the heading characteristics of the unmanned ship system, and constructs a dynamic model of each follower of the unmanned ship system based on the dynamic model of the heading characteristics of the unmanned ship system, including:

[0022] The dynamic model of the heading characteristics of the unmanned ship system introduced is shown in formula (4):

[0023]

[0024] Where: φ is the heading angle; τ is the actual control rudder angle; τ ω It represents the equivalent interference rudder angle caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; T represents the time constant, is the Norrbin coefficient; represents the first-order derivative operation, represents the second-order derivative operation;

[0025] According to the dynamic model of the heading characteristics of the unmanned ship system, a dynamic model of each follower of the unmanned ship system is constructed, as shown in formula (5):

[0026]

[0027] Among them: state variable x i,1=φ i,1 Indicates the heading angle state signal of the i-th follower unmanned ship, the state variable represents the state derivative signal of the heading angle of the i-th follower unmanned ship, represents the control input of the i-th follower unmanned ship system, ω i represents the equivalent interference, τ i is the actual control rudder angle of the i-th follower unmanned ship; τ ωi represents the equivalent interference rudder angle of the i-th follower unmanned ship caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; Τ represents the time constant, y i represents the output of the i-th follower unmanned boat.

[0028] Furthermore, the filter is specifically:

[0029] The filter is shown in formula (6),

[0030]

[0031] Where n = 1, 2, represents the dimension; μ i,n represents the fuzzy basis vector, η i,n The filtered blurred basis vectors, represents the parameter vector μ i,n The derivative of represents the filter parameters; x i,1 Indicates the heading angle state signal of the i-th follower unmanned ship, x i,2 represents the heading angle state derivative signal of the i-th follower unmanned ship, s i,n represents the filtered unmanned ship heading angle state signal, v i,n Indicates the actual unmanned ship heading angle state signal, θ i,n represents the filter gain, and t represents time.

[0032] Furthermore, S4 constructs a data-driven fuzzy estimator based on the fuzzy logic system, the dynamic model of each follower of the UAV system, and the data stored in the data stack, including:

[0033] The data-driven fuzzy predictor is constructed as shown in formulas (7), (8) and (9),

[0034]

[0035] Where: n i,1 is the preset performance constraint parameter, represents the estimated value of the preset performance constraint parameter; represents the estimation error of the preset performance constraint parameters, where Ci,0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, y i represents the output of the i-th follower unmanned boat, y j represents the output of the jth neighbor follower unmanned boat of the i-th follower unmanned boat; i,1 satisfy χ i,1 (0)>0,i=1,2,3,χ i,1 represents the finite-time preset performance function, π i,1 is with χ i,1 The associated locally Lipschitz continuous Class functions, is the comparison function, χ i,1 (∞) represents χ i,1 The final value of χ i,1 (0) represents χ i,1 The initial value of Denotes the finite time preset performance function χ i,1 The derivative of Satisfaction: g(ι i,1 )=0,ι i,1 ≥0, g(ι i,1 )=1,ι i,1 <0;q i,u ,q i,l is a positive scalar, g(ι i,1 ) represents a piecewise function; represents the degree of the jth neighbor leader drone ship associated with the i-th follower drone ship; a i,j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i,j =1; otherwise a i,j =0;a i,r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i,r =1; otherwise a i,r =0;x j,2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; y r represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i,1 , W i,2 represents the adaptive parameter, It's W i,1 The estimated value of It's W i,2 The estimated value of η i,1 , η i,2 represents the fuzzy basis vector; c i,1 represents the adjustment parameter of the preset performance constraint parameter fuzzy estimator; x i,2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship; It is x i,2 The estimated value of c i,2 Represents the heading angle state derivative signal x of the i-th follower unmanned ship i,2 The data drives the adjustment parameters of the fuzzy predictor; Δ i,1 , Δ i,2 Both represent the control gains of the data-driven fuzzy predictor; u i Represents the control input of the i-th follower unmanned ship system.

[0036] Furthermore, S5 introduces a virtual control law based on the dynamic surface, and constructs a second-order nonlinear tracking differentiator according to the virtual control law based on the dynamic surface, including:

[0037] The construction of the second-order nonlinear tracking differentiator is shown in formula (10),

[0038]

[0039] in: is the filtered virtual control law based on the dynamic surface, is the derivative of the virtual control law based on the dynamic surface after filtering; is the adjustment parameter of the second-order nonlinear tracking differentiator, represents the set of positive real numbers; α i,1 is a virtual control law based on dynamic surface; It represents the second-order derivative of the virtual control law based on the dynamic surface after filtering, that is, the output of the second-order nonlinear tracking differentiator.

[0040] Beneficial effects: The present invention provides a distributed finite-time data-driven heading consistency control method for unmanned ships, constructs a distributed finite-time consistency controller, and can adjust the convergence performance of synchronization error by adjusting a preset performance function, effectively suppressing the overshoot of synchronization error, improving the transient and steady-state performance of the system, and achieving rapid response;

[0041] The present invention adopts distributed cluster control in the controller design, which can achieve the control goal even when the computing resources are limited, the communication range is short, the communication bandwidth is narrow, and the intelligent agent to be manipulated and controlled is large;

[0042] The present invention designs a data-driven fuzzy predictor, relaxes the continuous excitation condition required for parameter convergence, and is more suitable for practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to 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 any creative labor.

[0044] Figure 1 A method flow chart of a distributed finite-time unmanned ship data-driven heading consistency control method provided by the present invention;

[0045] Figure 2 Schematic diagram of the structure of the distributed finite-time unmanned ship data-driven heading consistency controller designed for the present invention;

[0046] Figure 3 This is a communication topology diagram of an unmanned ship heading control system according to an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of synchronization error and finite time performance envelope according to an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of control input of an unmanned vessel system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0050] This embodiment provides a distributed finite-time unmanned ship data-driven heading consistency control method, such as Figure 1 Shown, including:

[0051] S1: Introducing a dynamic model of the unmanned ship system's heading characteristics, and constructing a dynamic model of each follower of the unmanned ship system based on the dynamic model of the unmanned ship system's heading characteristics, to obtain an unmanned ship heading angle state signal and an unmanned ship heading angle state derivative signal;

[0052] S2: Select a finite time preset performance function and obtain preset performance constraint parameters;

[0053] S3: introducing a fuzzy logic system, a filter and a data stack, filtering the unmanned ship heading angle state derivative signal and the fuzzy basis vector in the fuzzy logic system, obtaining filtered data and saving it through the data stack;

[0054] S4: constructing a data-driven fuzzy predictor based on the fuzzy logic system, the dynamic model of each follower of the unmanned ship system, and the data stored in the data stack, wherein the data-driven fuzzy predictor is used to receive the estimated value of the fuzzy weight sent by the fuzzy logic system and predict the approximate value of the unknown nonlinear function;

[0055] S5: introducing a virtual control law based on a dynamic surface, and constructing a second-order nonlinear tracking differentiator according to the virtual control law based on the dynamic surface; the second-order nonlinear tracking differentiator is used to receive the approximation value of the unknown nonlinear function and the virtual control law from the data-driven fuzzy predictor;

[0056] S6: constructing a distributed finite-time consistency controller according to the dynamic model of each follower of the unmanned ship, a preset performance function, preset performance constraint parameters, a data-driven fuzzy predictor, and a second-order nonlinear tracking differentiator to perform finite-time consistency control on the unmanned ship;

[0057] S7: Combining the fuzzy logic system, the second-order nonlinear tracking differentiator, the data stack, the filter, the data-driven fuzzy predictor, the unmanned ship system and the distributed finite-time consistency controller, a distributed finite-time unmanned ship data-driven heading consistency controller is obtained to achieve the unmanned ship heading consistency control.

[0058] Specifically, first, a dynamic model of the heading characteristics of the unmanned ship system is introduced, and a dynamic model of each follower of the unmanned ship system is constructed according to the dynamic model of the heading characteristics of the unmanned ship system, and the unmanned ship heading angle state signal and the unmanned ship heading angle state derivative signal are obtained, so as to prepare for the subsequent construction of a data-driven fuzzy predictor and controller for easy analysis; secondly, a finite time preset performance function is selected and preset performance constraint parameters are obtained, which can ensure that the tracking error does not exceed the preset value, effectively suppress the overshoot of the synchronization error, improve the transient and steady-state performance of the system, and achieve rapid response; a fuzzy logic system, a filter and a data stack are introduced to calculate the heading angle of the unmanned ship. The state derivative signal and the fuzzy basis vector in the fuzzy logic system are filtered to obtain the filtered data and save it through the data stack. The filter can filter the noisy unmanned ship navigation angle information caused by external disturbances to obtain more accurate unmanned ship navigation angle information; the fuzzy logic system is introduced, and a data-driven fuzzy predictor is constructed according to the fuzzy logic system, the dynamic model of each follower of the unmanned ship system and the data saved in the data stack. The data-driven fuzzy predictor is used to receive the estimated value of the fuzzy weight sent from the fuzzy logic system, predict the approximate value of the unknown nonlinear function, and design a data-driven fuzzy predictor that can relax the parameter convergence requirement. The continuous excitation condition of the unmanned ship is more in line with the actual application; a virtual control law based on the dynamic surface is introduced, and a second-order nonlinear tracking differentiator is constructed according to the virtual control law based on the dynamic surface; the second-order nonlinear tracking differentiator is used to receive the approximate value and virtual control law of the unknown nonlinear function from the data-driven fuzzy predictor, has a better differential tracking effect, is much less sensitive to noise than the classical differential tracker, and has better dynamic performance; a distributed finite-time consistency controller is constructed according to the dynamic model of each follower of the unmanned ship, a preset performance function, preset performance constraint parameters, a data-driven fuzzy predictor and a second-order nonlinear tracking differentiator to perform finite-time consistency control on the unmanned ship. Time consistency control adopts distributed cluster control, which can achieve control goals even when computing resources are limited, communication range is short, communication bandwidth is narrow, and the number of intelligent agents being manipulated and controlled is large. By adjusting the preset performance function to adjust the convergence performance of the synchronization error, the overshoot of the synchronization error can be effectively suppressed, the transient and steady-state performance of the system can be improved, and rapid response can be achieved. Finally, by combining fuzzy logic system, second-order nonlinear tracking differentiator, data stack, filter, data-driven fuzzy predictor, unmanned ship system and distributed finite-time consistency controller, a distributed finite-time unmanned ship data-driven heading consistency controller is obtained to realize the unmanned ship heading consistency control.

[0059] In a specific embodiment, a dynamic model of the unmanned ship system's heading characteristics is introduced, and a dynamic model of each follower of the unmanned ship system is constructed based on the dynamic model of the unmanned ship system's heading characteristics. The scheme for obtaining the unmanned ship's heading angle state signal and the unmanned ship's heading angle state derivative signal is:

[0060] The dynamic model of the heading characteristics of the introduced unmanned ship system is shown in formula (11):

[0061]

[0062] Where: φ is the heading angle; τ is the actual control rudder angle; τ ω It represents the equivalent interference rudder angle caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; T represents the time constant, is the Norrbin coefficient; represents the first-order derivative operation, represents the second-order derivative operation;

[0063] According to the dynamic model of the heading characteristics of the unmanned ship system, a dynamic model of each follower of the unmanned ship system is constructed, as shown in formula (12):

[0064]

[0065] Among them: state variable x i,1 =φ i,1 Indicates the heading angle state signal of the i-th follower unmanned ship, the state variable represents the state derivative signal of the heading angle of the i-th follower unmanned ship, represents the control input of the i-th follower unmanned ship system, ω i represents the equivalent interference, τ i is the actual control rudder angle of the i-th follower unmanned ship; τ ωi represents the equivalent interference rudder angle of the i-th follower unmanned ship caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; Τ represents the time constant, y i represents the output of the i-th follower unmanned boat.

[0066] In this scheme, a dynamic model of each follower of the unmanned ship system is constructed based on the dynamic model of the heading characteristics of the unmanned ship system, and the unmanned ship heading angle state signal and the unmanned ship heading angle state derivative signal are obtained, which prepares for the subsequent construction of data-driven fuzzy predictor and controller and facilitates analysis.

[0067] In a specific embodiment, the scheme for selecting a finite-time preset performance function and obtaining the preset performance constraint parameters is:

[0068] S21. Define heading tracking error;

[0069] S22. Constructing a preset performance boundary according to the heading tracking error, and selecting a finite-time preset performance function according to the preset performance boundary;

[0070] S23, constructing an error conversion function according to the heading tracking error and the finite time preset performance function;

[0071] S24. Construct preset performance constraint parameters according to the finite time preset performance function and the error conversion function.

[0072] This scheme selects a finite-time preset performance function and obtains preset performance constraint parameters, which can ensure that the tracking error does not exceed the preset value, effectively suppress the overshoot of the synchronization error, improve the transient and steady-state performance of the system, and achieve rapid response.

[0073] In a specific embodiment, a fuzzy logic system, a filter, and a data stack are introduced to filter the unmanned ship heading angle state derivative signal and the fuzzy basis vector in the fuzzy logic system, and the filtered data is obtained and saved through the data stack. The scheme is:

[0074] The filter is shown in formula (13),

[0075]

[0076] Where n = 1, 2, represents the dimension; μ i,n represents the fuzzy basis vector, η i,n The filtered blurred basis vectors, represents the parameter vector μ i,n The derivative of represents the filter parameters; x i,1 Indicates the heading angle state signal of the i-th follower unmanned ship, x i,2 represents the heading angle state derivative signal of the i-th follower unmanned ship, s i,n Represents the filtered unmanned ship heading angle state signal, υ i,n Indicates the actual unmanned ship heading angle state signal, θ i,n represents the filter gain, and t represents time.

[0077] In this scheme, a filter is constructed to filter the unmanned ship navigation angle derivative state signal with noise caused by external disturbances, so as to obtain a more accurate unmanned ship navigation angle state derivative signal.

[0078] In a specific embodiment, a data-driven fuzzy predictor is constructed based on the fuzzy logic system, the dynamic model of each follower of the unmanned vessel system, and the data stored in the data stack. The data-driven fuzzy predictor is used to receive the estimated value of the fuzzy weight sent by the fuzzy logic system. The scheme for predicting the approximate value of the unknown nonlinear function is:

[0079] The data-driven fuzzy predictor is constructed as shown in formulas (14), (15) and (16),

[0080]

[0081] Where: n i,1 is the preset performance constraint parameter, represents the estimated value of the preset performance constraint parameter; represents the estimation error of the preset performance constraint parameters, where C i,0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, y i represents the output of the i-th follower unmanned boat, y j represents the output of the jth neighbor follower unmanned boat of the i-th follower unmanned boat; i,1 satisfy χ i,1 (0)>0,i=1,2,3,χ i,1 represents the finite-time preset performance function, π i,1 is with χ i,1 The associated locally Lipschitz continuous Class functions, is the comparison function, χ i,1 (∞) represents χ i,1 The final value of χ i,1 (0) represents χ i,1 The initial value of Denotes the finite time preset performance function χ i,1 The derivative of Satisfaction: g(ι i,1 )=0,ι i,1 ≥0, g(ι i,1 )=1,ι i,1 <0;q i,u ,q i,l is a positive scalar, g(ι i,1 ) represents a piecewise function; represents the degree of the jth neighbor leader drone ship associated with the i-th follower drone ship; a i,j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i,j =1; otherwise a i,j =0;a i,r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i,r =1; otherwise ai,r =0;x j,2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; y r represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i,1 , W i,2 represents the adaptive parameter, It's W i,1 The estimated value of It's W i,2 The estimated value of η i,1 , η i,2 represents the fuzzy basis vector; c i,1 represents the adjustment parameter of the preset performance constraint parameter fuzzy estimator; x i,2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship; It is x i,2 The estimated value of c i,2 Represents the heading angle state derivative signal x of the i-th follower unmanned ship i,2 The data drives the adjustment parameters of the fuzzy predictor; Δ i,1 , Δ i,2 Both represent the control gains of the data-driven fuzzy predictor; u i Represents the control input of the i-th follower unmanned ship system.

[0082] In this scheme, a data-driven fuzzy predictor is designed to relax the continuous excitation condition required for parameter convergence, which is more in line with practical applications.

[0083] In a specific embodiment, a virtual control law based on a dynamic surface is introduced, and a second-order nonlinear tracking differentiator is constructed according to the virtual control law based on the dynamic surface; the second-order nonlinear tracking differentiator is used to receive the approximation value of the unknown nonlinear function from the data-driven fuzzy predictor and the virtual control law.

[0084] The construction of the second-order nonlinear tracking differentiator is shown in formula (17),

[0085]

[0086] in: is the filtered virtual control law based on the dynamic surface, is the derivative of the virtual control law based on the dynamic surface after filtering; is the adjustment parameter of the second-order nonlinear tracking differentiator, represents the set of positive real numbers; α i,1 is a virtual control law based on dynamic surface; It represents the second-order derivative of the virtual control law based on the dynamic surface after filtering, that is, the output of the second-order nonlinear tracking differentiator.

[0087] In this scheme, a second-order nonlinear tracking differentiator is constructed. Compared with the linear tracking differentiator, it has better differential tracking effect, is much less sensitive to noise than the classic differential tracker, and has better dynamic performance.

[0088] In a specific embodiment, a distributed finite-time consistency controller is constructed based on the dynamic model of each follower of the unmanned ship, a preset performance function, preset performance constraint parameters, a data-driven fuzzy predictor, and a second-order nonlinear tracking differentiator. The scheme for finite-time consistency control of the unmanned ship is:

[0089] The distributed finite-time consistency controller is shown in formulas (18), (19) and (20),

[0090]

[0091] Where: α i,1 is the virtual control law based on the dynamic surface, d i represents the sum of the degrees between the follower and its neighbors, represents the degree of the jth neighbor leader associated with the i-th follower drone ship; a i,j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i,j =1; otherwise a i,j =0;a i,r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i,r =1; otherwise a i,r =0;Δ i,1 , Δ i,2 represents the control gain of the data-driven fuzzy predictor; o i,1 represents the design parameters, C i,0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, χ i,1 Preset performance function for finite time, Represents the preset performance function χ i,1 The derivative of n i,1 is the preset performance constraint parameter; x j,2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; yr represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i,1 , W i,2 represents the adaptive parameter, It's W i,1 The estimated value of It's W i,2 The estimated value of η i,1 , η i,2 represents the blurred basis vector after filtering; n i,2 represents the dynamic surface error, x i,2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship, It is x i,2 The estimated value of represents the estimation error; represents the filtered virtual control law based on the dynamic surface; is the adaptive gain; is the adaptive parameter; It means t=t p Time μ i,2 and ν i,2 The value of Indicates the stack length; represents the filtered second-order nonlinear tracking differentiator; for The transpose of W i,2 The transpose of u i represents the control input of the i-th follower unmanned ship system; degree refers to the number of edges connected to the intelligent agent in the communication topology, and in this embodiment, the number of edges connected to the data in the communication topology related to the unmanned ship.

[0092] In this scheme, a finite-time unmanned ship heading consistency controller is constructed, which adopts distributed cluster control. It can achieve the control target even when computing resources are limited, communication range is short, communication bandwidth is narrow, and the intelligent agent to be manipulated and controlled is large. In the controller, the convergence performance of the synchronization error can be adjusted by adjusting the preset performance function, which can effectively suppress the overshoot of the synchronization error, improve the transient and steady-state performance of the system, and achieve rapid response.

[0093] In a specific embodiment, a fuzzy logic system, a second-order nonlinear tracking differentiator, a data stack, a filter, a data-driven fuzzy predictor, an unmanned ship system, and a distributed finite-time consistency controller are combined to obtain a distributed finite-time unmanned ship data-driven heading consistency controller. The solution for achieving the unmanned ship heading consistency control is:

[0094] The fuzzy logic system, the second-order nonlinear tracking differentiator, the data stack, the filter, the data-driven fuzzy predictor, the unmanned ship system and the distributed finite-time consistency controller are combined to form a distributed finite-time unmanned ship data-driven heading consistency controller, such as Figure 2 As shown, the output of the unmanned vessel system is connected to the input of the filter, the output of the filter is connected to the input of the data stack, the output of the data stack is connected to the input of the fuzzy logic system, and the input and output of the fuzzy logic system are simultaneously connected to the data-driven fuzzy predictor. The output of the data-driven fuzzy predictor is connected to the input of the second-order nonlinear tracking differentiator and the input of the distributed finite-time consistency controller, respectively. The output of the second-order nonlinear tracking differentiator is connected to the input of the distributed finite-time consistency controller, and the output of the distributed finite-time consistency controller is connected to the input of the unmanned vessel system.

[0095] Specifically, the filter receives a heading angle state signal and an unmanned ship heading angle state derivative signal from the unmanned ship system, and sends the filtered information to a data stack;

[0096] The data stack receives the filtered information from the filter and sends the filtered information to the fuzzy logic system;

[0097] The fuzzy logic system receives information stored in the data stack and generates an error value between an estimated value of the unmanned ship heading angle state derivative signal and the unmanned ship system heading angle state derivative signal, driven by the data;

[0098] The data-driven fuzzy predictor receives the estimated value of the fuzzy weight from the fuzzy logic system, sends the approximation value of the unknown nonlinear function and the virtual control law to the second-order nonlinear tracking differentiator, and sends the approximation value of the unknown nonlinear function to the distributed finite-time consistency controller;

[0099] The second-order nonlinear tracking differentiator receives the approximation value of the unknown nonlinear function and the virtual control law from the data-driven fuzzy predictor, and sends the second-order derivative information of the virtual control law to the distributed finite-time consistency controller;

[0100] The distributed finite-time consistency controller receives a given reference heading signal, an approximation of an unknown nonlinear function from a data-driven fuzzy predictor, and second-order derivative information of a virtual control law obtained by a second-order nonlinear tracking differentiator, and sends the actual control law to the unmanned ship system.

[0101] The unmanned ship system receives the actual control law and given external environmental disturbance from the distributed finite-time consistency controller, sends the unmanned ship heading angle state signal and the unmanned ship heading angle state derivative signal to the filter, completes the iterative process, and realizes the unmanned ship heading consistency control.

[0102] An embodiment of the present invention is as follows:

[0103] At time t, first, for a given reference heading signal y r , using a distributed finite-time consensus controller to generate the control input u i , then, control input u i , external disturbance is applied to the unmanned ship system and the resulting unmanned ship heading angle x i,1 And its derivative information x i,2 The information is sent to the filter for filtering, and then the filtered information is sent to the data stack for storage; secondly, the information stored in the data stack and the estimated error of the heading angle derivative information are The estimated values ​​of the preset performance constraint parameters are sent to the fuzzy logic system, and the adaptive parameter estimates generated by the fuzzy logic system are The signal generated by the data-driven fuzzy predictor and the virtual control law are sent to the second-order nonlinear tracking differentiator and the distributed finite-time consistency controller respectively; the distributed finite-time consistency controller is based on the second-order derivative of the filtered virtual control law generated by the second-order nonlinear tracking differentiator and the given reference signal y r The information generated by the data-driven fuzzy predictor generates the control signal for the next moment, thereby completing the control iteration process and realizing the heading consistency control of the unmanned ship.

[0104] The control parameters for the embodiment are selected as follows:

[0105] T=0.707,L=1.102, q i,u =1,q i,l =0.8,

[0106]

[0107] π i,1 =0.35,χ i,1 (∞)=0.03.

[0108] The simulation communication topology diagram is as follows Figure 3 As shown, according to Figure 3 The communication relationship is used for communication, and the simulation results are as follows Figure 4-5 As shown, Figure 4 The synchronization error and performance envelope diagram of the distributed finite time unmanned ship data driven heading consistency of the present invention are given. It can be seen that the synchronization error It can converge to the preset performance boundary within a limited time; Figure 5 The control input diagram of the unmanned ship heading consistency control system of the present invention is given. It can be seen that the control input ui It can converge to the neighborhood of the origin. Therefore, by implementing the proposed controller for control, the synchronization error can converge to a small residual set, which proves that the control method of the present invention can achieve the heading consistency control of the unmanned ship.

[0109] 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 above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above 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 distributed finite-time data-driven heading consistency control method for unmanned ships, characterized by: include: S1: Introducing a dynamic model of the unmanned ship system's heading characteristics, and constructing a dynamic model of each follower of the unmanned ship system based on the dynamic model of the unmanned ship system's heading characteristics, to obtain an unmanned ship heading angle state signal and an unmanned ship heading angle state derivative signal; S2: Select a finite time preset performance function and obtain preset performance constraint parameters; S3: introducing a fuzzy logic system, a filter and a data stack, filtering the unmanned ship heading angle state derivative signal and the fuzzy basis vector in the fuzzy logic system, obtaining filtered data and saving it through the data stack; S4: constructing a data-driven fuzzy predictor based on the fuzzy logic system, the dynamic model of each follower of the unmanned ship system, and the data stored in the data stack, wherein the data-driven fuzzy predictor is used to receive the estimated value of the fuzzy weight sent by the fuzzy logic system and predict the approximate value of the unknown nonlinear function; S5: introducing a virtual control law based on a dynamic surface, and constructing a second-order nonlinear tracking differentiator according to the virtual control law based on the dynamic surface; the second-order nonlinear tracking differentiator is used to receive the approximation value of the unknown nonlinear function and the virtual control law from the data-driven fuzzy predictor; S6: constructing a distributed finite-time consistency controller according to the dynamic model of each follower of the unmanned ship, a preset performance function, preset performance constraint parameters, a data-driven fuzzy predictor, and a second-order nonlinear tracking differentiator to perform finite-time consistency control on the unmanned ship; S7: Combining the fuzzy logic system, the second-order nonlinear tracking differentiator, the data stack, the filter, the data-driven fuzzy predictor, the unmanned ship system and the distributed finite-time consistency controller, a distributed finite-time unmanned ship data-driven heading consistency controller is obtained to achieve the unmanned ship heading consistency control.

2. A distributed finite-time unmanned ship data-driven heading consistency control method according to claim 1, characterized in that: S6 builds a distributed finite-time consistency controller, including: The distributed finite-time consistency controller is shown in formulas (1), (2) and (3), Where: α i,1 is the virtual control law based on the dynamic surface, d i represents the sum of the degrees between the follower and its neighbors, represents the degree of the jth neighbor leader associated with the i-th follower drone ship; a i，j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i,j =1; otherwise a i,j =0;a i，r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i，r =1; otherwise a i，r =0;Δ i，1 , Δ i，2 represents the control gain of the data-driven fuzzy predictor; o i，1 represents the design parameters, C i，0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, χ i，1 Preset performance function for finite time, Represents the preset performance function χ i，1 The derivative of n i，1 is the preset performance constraint parameter; x j，2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; y r represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i，1 , W i，2 represents the adaptive parameter, It's W i，1 The estimated value of It's W i，2 The estimated value of η i，1 , η i，2 represents the blurred basis vector after filtering; n i，2 represents the dynamic surface error, x i，2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship, It is x i，2 The estimated value of represents the estimation error; represents the filtered virtual control law based on the dynamic surface; is the adaptive gain; is the adaptive parameter; It means t=t p Time μ i，2 and v i，2 The value of Indicates the stack length; represents the filtered second-order nonlinear tracking differentiator; for The transpose of W i，2 The transpose of u i Represents the control input of the i-th follower unmanned ship system.

3. The distributed finite-time data-driven heading consistency control method for unmanned ships according to claim 1 is characterized in that: S1 introduces a dynamic model of the heading characteristics of the unmanned ship system, and constructs a dynamic model of each follower of the unmanned ship system based on the dynamic model of the heading characteristics of the unmanned ship system, including: The dynamic model of the heading characteristics of the unmanned ship system introduced is shown in formula (4): Where: φ is the heading angle; τ is the actual control rudder angle; τ ω It represents the equivalent interference rudder angle caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; T represents the time constant, is the Norrbin coefficient; represents the first-order derivative operation, represents the second-order derivative operation; According to the dynamic model of the heading characteristics of the unmanned ship system, a dynamic model of each follower of the unmanned ship system is constructed, as shown in formula (5): Among them: state variable x i，1 =φ i，1 Indicates the heading angle state signal of the i-th follower unmanned ship, the state variable represents the state derivative signal of the heading angle of the i-th follower unmanned ship, represents the control input of the i-th follower unmanned ship system, ω i represents the equivalent interference, τ i is the actual control rudder angle of the i-th follower unmanned ship; τ ωi represents the equivalent interference rudder angle of the i-th follower unmanned ship caused by environmental disturbances such as wind, waves and ocean currents; L represents the gain constant; T represents the time constant, y i represents the output of the i-th follower unmanned boat.

4. The distributed finite-time data-driven heading consistency control method for unmanned ships according to claim 1 is characterized in that: The filter is specifically: The filter is shown in formula (6), Where n = 1, 2, represents the dimension; μ i，n represents the fuzzy basis vector, η i，n The filtered blurred basis vectors, represents the parameter vector μ i，n The derivative of represents the filter parameters; x i，1 Indicates the heading angle state signal of the i-th follower unmanned ship, x i，2 represents the heading angle state derivative signal of the i-th follower unmanned ship, s i，n Represents the filtered unmanned ship heading angle state signal, υ i，n Indicates the actual unmanned ship heading angle state signal, θ i，n represents the filter gain, and t represents time.

5. The distributed finite-time data-driven heading consistency control method for unmanned ships according to claim 3 is characterized in that: S4 builds a data-driven fuzzy estimator based on the fuzzy logic system, the dynamic model of each follower of the UAV system, and the data stored in the data stack, including: The data-driven fuzzy predictor is constructed as shown in formulas (7), (8) and (9), Where: n i，1 is the preset performance constraint parameter, represents the estimated value of the preset performance constraint parameter; represents the estimation error of the preset performance constraint parameters, where o i，1 =C i，0 ι i，1 l i，1 / χ i，1 , C i，0 ∈(0,1) is the design constant, is the error transformation function, is the synchronization error, y i represents the output of the i-th follower unmanned boat, y j represents the output of the jth neighbor follower unmanned boat of the i-th follower unmanned boat; i，1 satisfy χ i，1 (0)>0, i=1, 2, 3, χ i，1 represents the finite-time preset performance function, π i，1 is with χ i，1 The associated locally Lipschitz continuous Class functions, is the comparison function, χ i，1 (∞) represents χ i，1 The final value of χ i，1 (0) represents χ i，1 The initial value of Denotes the finite time preset performance function χ i，1 The derivative of Satisfaction: g(ι i，1 )=0,ι i，1 ≥0, g(ι i，1 )=1,ι i，1 <0;q i，u ,q i，l is a positive scalar, g(ι i，1 ) represents a piecewise function; represents the degree of the jth neighbor leader drone ship associated with the i-th follower drone ship; a i，j Indicates whether there is a communication relationship between the i-th follower unmanned boat and the j-th neighbor leader unmanned boat. If the i-th follower unmanned boat can obtain the information of the j-th neighbor leader unmanned boat, then a i，j =1; otherwise a i，j =0;a i，r Indicates whether there is a communication relationship between the i-th follower unmanned ship and the r-th neighbor follower unmanned ship. If the i-th follower unmanned ship can obtain the information of the r-th neighbor follower unmanned ship, then a i，r =1; otherwise a i，r =0;x j，2 represents the state derivative signal of the heading angle of the neighbor follower unmanned ship of the i-th follower unmanned ship; y r represents the motion dynamics of the i-th follower unmanned boat, represents y r The derivative of W i，1 , W i，2 represents the adaptive parameter, It's W i，1 The estimated value of It's W i，2 The estimated value of η i，1 , η i，2 represents the fuzzy basis vector; c i，1 represents the adjustment parameter of the preset performance constraint parameter fuzzy estimator; x i，2 represents the state derivative signal of the heading angle of the i-th follower unmanned ship; It is x i，2 The estimated value of c i，2 Represents the heading angle state derivative signal x of the i-th follower unmanned ship i，2 The data drives the adjustment parameters of the fuzzy predictor; Δ i，1 , Δ i，2 Both represent the control gains of the data-driven fuzzy predictor; u i Represents the control input of the i-th follower unmanned ship system.

6. The distributed finite-time data-driven heading consistency control method for unmanned ships according to claim 1 is characterized in that: S5 introduces a virtual control law based on the dynamic surface, and constructs a second-order nonlinear tracking differentiator according to the virtual control law based on the dynamic surface, including: The construction of the second-order nonlinear tracking differentiator is shown in formula (10), in: is the filtered virtual control law based on the dynamic surface, is the derivative of the virtual control law based on the dynamic surface after filtering; is the adjustment parameter of the second-order nonlinear tracking differentiator, represents the set of positive real numbers; α i，1 is a virtual control law based on dynamic surface; It represents the second-order derivative of the virtual control law based on the dynamic surface after filtering, that is, the output of the second-order nonlinear tracking differentiator.

Images