Unmanned ship adaptive fuzzy course tracking control method with input quantization, state quantization and input and output constraints

Abstract

The invention provides an unmanned ship self-adaptive fuzzy course tracking control method with input quantization, state quantization and input and output constraints, and the method comprises the steps: carrying out the input quantization and state quantization through a uniform quantizer, and carrying out the linear description of a quantization process, and clarifying the quantization process; a logarithmic obstacle Lyapunov function is introduced, and it is guaranteed that the state of the system is kept within a reasonable range; uncertainty items in the ship model are estimated through a fuzzy logic system, and the influence of external disturbance on a control system is reduced; based on the Lyapunov stability theory and the proposed lemma, the bounded quantization error and the stability of the designed control strategy are proved. According to the method, the logarithmic obstacle function and the fuzzy logic system are introduced, the Lyapunov stability theory is combined, stability analysis is carried out on the designed control system, output constraints are met, error signals of a closed-loop system are consistent and finally bounded, and accurate tracking of the USV course and guarantee of the system stability can be achieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an unmanned ship adaptive fuzzy heading tracking control method with input quantization, state quantization and input-output constraints. Background Art

[0002] With the rapid development of modern automation technology and unmanned systems, USVs have shown great application potential and value in many fields such as operations in dangerous or complex marine environments, marine resource exploration, and pollution monitoring, making them one of the current research hotspots in the field of marine engineering.

[0003] The heading control of unmanned ships is one of the most important research directions in the field of ship motion control. In order to achieve precise heading tracking control, many novel and valuable control methods have been applied to the heading control of unmanned ships, such as adaptive fuzzy control, neural network control, sliding mode control, model predictive control (MPC), PID control and other heading control methods.

[0004] In the marine environment of unmanned vessels, their actual position cannot change freely and must be adjusted or controlled within a certain spatial or temporal range. Position constraints involve multiple factors, including the natural environment (such as shoals and tides), traffic management (such as waterways and restricted areas), and equipment limitations (such as the accuracy of navigation equipment). Understanding and adhering to these constraints is key to ensuring navigation safety and efficiency. Position constraints are quite common in practical applications, and researchers have conducted extensive research on state or output constraints. However, the existing problem of unmanned vessel heading tracking control with output constraints remains poorly addressed. In recent years, the barrier Lyapunov function (BLF) has played an increasingly important role in the field of constrained control, providing an effective method to ensure that the system remains stable and meets performance requirements even in the presence of constraints. Existing research has not considered combining input quantization with state quantization and input-output constraints in heading tracking control. To improve the safety, robustness, and stability of control systems, control systems that combine input quantization, state quantization, input constraints, and output constraints are of great significance in the field of navigation.

[0005] In practice, USVs operate in a networked environment, and information between controllers and actuators is typically transmitted via wireless communication. Due to limited communication bandwidth at sea, quantitative control can reduce the rate within the bandwidth and address the limited network bandwidth, optimizing control performance and reducing power consumption. At sea, USVs often need to respond to sudden changes, such as waves and wind speed. Quantized control enables faster parameter adjustments and improved response speed, ensuring that the heading control system can adapt to environmental changes in real time and maintain a stable heading. The two most commonly used quantizers for quantitative control are uniform quantizers and hysteresis quantizers, which mitigate the limitations of maritime communication bandwidth. However, the introduction of quantization inevitably introduces quantization error. Excessive quantization error can affect the stability of the control system and its heading tracking capability. In existing research, quantization error is typically analyzed and treated as a disturbance of an unquantized variable, and appropriate error compensation techniques are developed by estimating the boundedness of the quantization error. However, the effectiveness of these research methods is significantly reduced if the quantization parameters are time-varying. Summary of the Invention

[0006] Based on the technical problems raised above, a method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints is provided. The present invention introduces a logarithmic barrier function and a fuzzy logic system, combines the Lyapunov stability theory, and performs stability analysis on the designed control system. The output constraints are satisfied, and the closed-loop system error signal is consistent and ultimately bounded, which can achieve accurate tracking of the USV heading and ensure system stability.

[0007] The technical means adopted in the present invention are as follows:

[0008] An adaptive fuzzy heading tracking control method for an unmanned ship with input quantization, state quantization and input-output constraints comprises:

[0009] S1. The uniform quantizer is used to quantize the input and state, and the quantization process is described linearly to illustrate the quantization process.

[0010] S2. Introduce the logarithmic barrier Lyapunov function to ensure that the system state remains within a reasonable range;

[0011] S3. Use fuzzy logic system to estimate uncertainty terms in the ship model and reduce the impact of external disturbances on the control system;

[0012] S4. Based on Lyapunov stability theory and the proposed lemma, the boundedness of the quantization error and the stability of the designed control strategy are proved.

[0013] Furthermore, step S1 specifically includes:

[0014] S11. Establish a mathematical model for USV heading control as follows:

[0015]

[0016] Among them, ψ represents the ship's heading angle; r represents the ship's yaw rate; β represents the disturbance caused by factors such as wind, waves, and currents; τ represents the control input of the system; b represents the control system gain, Where K represents the ship's turning index, T represents the ship's following index; a1 and a2 both represent the nonlinear coefficients of the Norrbin motion model.

[0017] S12. Define the state variables as ψ and r, let ψ = x1, r = x2, and establish the mathematical model of quantized USV heading control as follows:

[0018]

[0019] Where Q(τ) is the quantized control input, f(x,t)=a1r+a2r 3 ,The quantization process is to convert a continuous signal into a discrete signal;

[0020] S13. Assuming external disturbance and is a positive constant;

[0021] S14. Design a uniform quantizer for the state variables x1, x2 and the control input Quantization is performed to obtain Q(g), as follows:

[0022]

[0023] in, In the above formula, χ represents the quantization step size, and χ>0, L i =χ,L i+1 =L i +χ, i∈R;, quantization error will be generated during the quantization process, and the quantization error satisfies

[0024] S15. Let Q(τ) = q1(t)τ + q2(t), then:

[0025]

[0026] Among them, q1(t) is unknown and its sign remains unchanged during the quantization process. From the above formula, we can see that q1(t)>0. When |τ(t)|<a, Q(τ(t)) is also bounded. From this, we can infer that q2(t) is bounded.

[0027] Furthermore, step S2 specifically includes:

[0028] S21. Introduce input constraints in the controller and use the saturation function sat(τ) instead of the sign function sgn(τ) as follows:

[0029]

[0030] S22. Introduce the logarithmic barrier Lyapunov function, as follows:

[0031] Setting Lemma 1: For the error system where g = [z1z2] T , there exist continuously differentiable and positive definite functions V1 and V2; there exist two positive constants k a1 and k b1 , the position output is x1, and the heading signal error z1 is defined as x1-x 1d , when -k is satisfied a1 <z1<k b1 When V1(z1)→∞, γ1(||z2||)≤V2(z2)≤γ2(||z2||), γ1 and γ2 are K ∞ Class function; when -k a1 <z1(0)<k b1 , take V(z)=V1(z1)+V2(z2), if but z1(t)∈(-k a1 ,k b1 ), where λ and μ are constants;

[0032] Lemma 2: For k b1 >0, if full Then there is

[0033] On the error surface z1=x1-x 1d , z2=x2-α, α is a stable function that has not yet been designed. In order to achieve Define the logarithmic barrier function as follows:

[0034]

[0035] Where log(·) is the natural logarithm, k b1 =k d -B1 represents the constraint on z1, and we get |z1|<k b1 , and it can also be concluded that V1 is in |z1|<k b1 is positive definite, and taking the derivative of V1, we get:

[0036]

[0037] Design a stable function α as follows:

[0038]

[0039] Where c1 is a positive constant, and the derivative of the stable function is obtained:

[0040]

[0041] Substituting the stability function into In the equation, we get:

[0042]

[0043] When z2 is 0, then

[0044] Since x2 does not need to be constrained, V1 is augmented using a quadratic function and a Lyapunov function is designed as follows:

[0045] V3=V1+V2

[0046] in, Taking the derivative of the Lyapunov function V3, we get:

[0047]

[0048] Let F(X)=f(x,t)+β and Then we can get:

[0049]

[0050] Furthermore, step S3 specifically includes:

[0051] S31, F(X) is unknown, according to the function approximation theorem, using the fuzzy logic system θ *T ε(j) approximates the unfamiliar continuous function F(X). According to step S22, a fuzzy logic system is designed as follows:

[0052] F(X)=θ *T ε(j)+ω

[0053] in, represents the ideal weight, represents the fuzzy basis vector, ω represents the approximation error of the fuzzy logic system, and F(X) is a bounded function;

[0054] S32. Calculate the estimated value of the bounded function F(X). The calculation formula is as follows:

[0055]

[0056] in, is θ * An estimate of Make and

[0057] S33, Setting Note 1: Define time-varying gain n = 1 / q1(t) min , where q1(t) min It is the lower bound of q1(t). Since the value of q1(t) is unknown and time-varying, an adaptive method is used to estimate their boundaries and estimate the lower bound of q1(t), thereby avoiding the singular problem when it is estimated to be zero.

[0058] Furthermore, step S4 specifically includes:

[0059] S41. Design the adaptive fuzzy non-quantized control input signal and adaptive law as follows:

[0060]

[0061] Among them, there is an adaptive law and γ1,γ2,p,η,k2,c2,c,λ1 are all positive constants greater than zero, represents an estimate of n, and

[0062] S42, control input signal And the state variables, intermediate signals, and adaptive laws in the system are quantified as follows:

[0063]

[0064]

[0065] S43. Consider input quantization but not state quantization, and prove stability;

[0066] S44. Prove that the quantization error is bounded;

[0067] S45. Consider both input quantization and state quantization to prove stability.

[0068] Furthermore, step S43 specifically includes:

[0069] S431. Assume Lemma 3: Under the premise of considering the USV heading tracking control problem of Lemma 1 and Lemma 2, together with the control input, the adaptive law, the closed-loop control system error signal is consistent and ultimately bounded;

[0070] S432. Design the Lyapunov function as follows:

[0071]

[0072] The time derivative of the above equation is as follows:

[0073]

[0074] Substituting the control law into the above formula, we get:

[0075]

[0076] Substituting the adaptive law into the above formula, we get:

[0077]

[0078] S433、Set because so And because the time-varying gain n=1 / q1(t) min , so we introduce:

[0079]

[0080] Substituting into the above formula, we get:

[0081]

[0082] By the following formula:

[0083]

[0084] -λ1z2sgnz2≤-λ1|z2|

[0085] Then it can be deduced that:

[0086]

[0087] By Lemma 2, we get Then we get:

[0088]

[0089] S434, Order According to Lemma 1 and Lyapunov stability theory, the error of the USV heading tracking control system is bounded when state quantization is not considered, and the closed-loop system signal is ultimately bounded; and z1, z2, Bounded and obtainable It satisfies |z1|<k b ,t∈(0,∞); when t tends to infinity, V(t) converges to And its convergence accuracy depends on c. When c is much larger than λ, V(t)→0,z1→0,z2→0.

[0090] Furthermore, step S44 specifically includes:

[0091] S441. Set Lemma 4: Define the error surface, control input, state variable, and quantization error of the intermediate signal as

[0092] S442, from steps S22 and S42, β α 、 and Bounded, as follows:

[0093]

[0094] S443. From Lemma 3, we can see that the closed-loop system is stable when state quantization is not considered, and the error signal of the closed-loop control system is consistent and ultimately bounded. It is bounded.

[0095] S444: Launch according to step S41 According to step S22, it is deduced:

[0096]

[0097] And C and ν are both positive constants;

[0098] S445. According to step S41 and step S42, we have:

[0099]

[0100] S446: According to steps S41 and S42, β τ Bounded and combined with the formula in step S445, we can know that:

[0101]

[0102] Then there is a constant Make It can be seen from this that the quantization error in the designed quantitative feedback control system is bounded;

[0103] S447, Set Note 2: Because is a positive constant; z1, z2, is bounded, then we have is bounded, and o2 and δ are both bounded constants, where:

[0104]

[0105] S448. Considering the quantized state variables, intermediate signal formulas and control laws according to step S42, it is proved that the designed USV heading tracking control closed-loop system with quantization is stable, and the closed-loop system signals are consistent and ultimately bounded.

[0106] Furthermore, step S45 specifically includes:

[0107] S451. Design the time derivative of the Lyapunov function as follows:

[0108]

[0109] S452. According to step S44, the state quantization error of the designed quantization feedback system is bounded, and Then launch:

[0110]

[0111] S453, Order According to Lemma 1 and Lyapunov stability theory, it is proved that the USV heading tracking control closed-loop system with state quantization and input quantization is stable, and the closed-loop system signal is consistent and ultimately bounded.

[0112] Compared with the prior art, the present invention has the following advantages:

[0113] This invention provides an adaptive fuzzy heading tracking control method for an unmanned vessel with input quantization, state quantization, and input-output constraints. This method combines input-output constraints with state quantization and input quantization for heading tracking control. This method ensures that control signals are within the physical range and prevents control commands from exceeding practical operational limits. Setting constraints helps the system avoid instability or uncontrollability caused by inputs outside a reasonable range. In practical operating environments, constraints can enhance the system's robustness to uncertainty and disturbances.

[0114] 2. Compared with previous quantitative control methods, which assume constant input quantization parameters and treat quantization errors as disturbances in the control system, this invention uses a linear analytical model to describe input quantization. This allows the control law to be designed without prior knowledge of the input quantization parameters, thus enhancing the control system's adaptability to time-varying quantization parameters.

[0115] 3. Compared with the existing USV control results that rely on continuous feedback signals, the present invention takes into account the limitations of maritime communication bandwidth when sailing at sea and designs a ship heading tracking control strategy based on signal quantization, which reduces the transmission burden of maritime communication signals and reduces energy consumption.

[0116] Based on the above reasons, the present invention can be widely promoted in fields such as artificial intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0118] Figure 1 Flow chart of the method of the present invention.

[0119] Figure 2 The tracking results of the "Lanxin" USV provided in this embodiment of the present invention.

[0120] Figure 3 The tracking error of the "Lanxin" USV provided in an embodiment of the present invention.

[0121] Figure 4 This is the fuzzy approximation result provided by the embodiment of the present invention.

[0122] Figure 5 This is the control input for the "Lanxin" USV provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0123] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0124] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or apparatuses.

[0125] like Figure 1 As shown, the present invention provides an unmanned ship adaptive fuzzy heading tracking control method with input quantization, state quantization and input-output constraints, including:

[0126] S1. The uniform quantizer is used to quantize the input and state, and the quantization process is described linearly to illustrate the quantization process.

[0127] S2. Introduce the logarithmic barrier Lyapunov function to ensure that the system state remains within a reasonable range;

[0128] S3. Use fuzzy logic system to estimate uncertainty terms in the ship model and reduce the impact of external disturbances on the control system;

[0129] S4. Based on Lyapunov stability theory and the proposed lemma, the boundedness of the quantization error and the stability of the designed control strategy are proved.

[0130] In specific implementation, as a preferred embodiment of the present invention, step S1 specifically includes:

[0131] S11. Establish a mathematical model for USV heading control as follows:

[0132]

[0133] Among them, ψ represents the ship's heading angle; r represents the ship's yaw rate; β represents the disturbance caused by factors such as wind, waves, and currents; τ represents the control input of the system; b represents the control system gain, Where K represents the ship's turning index, T represents the ship's following index; a1 and a2 both represent the nonlinear coefficients of the Norrbin motion model.

[0134] S12. Define the state variables as ψ and r, let ψ = x1, r = x2, and establish the mathematical model of quantized USV heading control as follows:

[0135]

[0136] Where Q(τ) is the quantized control input, f(x,t)=a1r+a2r 3 ,The quantization process is to convert a continuous signal into a discrete signal;

[0137] S13. Assuming external disturbance and is a positive constant;

[0138] S14. Design a uniform quantizer for the state variables x1, x2 and the control input Quantization is performed to obtain Q(g), as follows:

[0139]

[0140] in, In the above formula, χ represents the quantization step size, and χ>0, Li =χ,L i+1 =L i +χ, i∈R;, quantization error will be generated during the quantization process, and the quantization error satisfies

[0141] S15. Let Q(τ) = q1(t)τ + q2(t), then:

[0142]

[0143] Among them, q1(t) is unknown and its sign remains unchanged during the quantization process. From the above formula, we can see that q1(t)>0. When |τ(t)|<a, Q(τ(t)) is also bounded. From this, we can infer that q2(t) is bounded.

[0144] In this embodiment, due to the limited capacity of the maritime network communication channel, quantization control is adopted when considering the heading tracking problem. Input quantization and state quantization are introduced, and a uniform quantizer is used for quantization.

[0145] In specific implementation, as a preferred embodiment of the present invention, step S2 specifically includes:

[0146] S21. In the controller, if the interference is too large, the designed switching gain λ1 needs to be larger, but this will cause larger chattering. To prevent the impact of chattering on the controller, an input constraint is introduced into the controller, and the saturation function sat(τ) is used instead of the sign function sgn(τ), as follows:

[0147]

[0148] S22. Introduce the logarithmic barrier Lyapunov function, as follows:

[0149] Setting Lemma 1: For the error system where g = [z1z2] T , there exist continuously differentiable and positive definite functions V1 and V2; there exist two positive constants k a1 and k b1 , the position output is x1, and the heading signal error z1 is defined as x1-x 1d , when -k is satisfied a1 <z1<k b1 When V1(z1)→∞, γ1(||z2||)≤V2(z2)≤γ2(||z2||), γ1 and γ2 are K ∞ Class function; when -k a1 <z1(0)<k b1 , take V(z)=V1(z1)+V2(z2), if but z1(t)∈(-k a1 ,k b1 ), where λ and μ are constants;

[0150] Lemma 2: For k b1 > 0, if |z1(t)| < k b1 , Then there is

[0151] On the error surface z1=x1-x 1d , z2=x2-α, α is a stable function that has not yet been designed. In order to achieve Define the logarithmic barrier function as follows:

[0152]

[0153] Where log(·) is the natural logarithm, k b1 =k d -B1 represents the constraint on z1, and we get |z1|<k b1 , and it can also be concluded that V1 is in |z1|<k b1 is positive definite, and taking the derivative of V1, we get:

[0154]

[0155] Design a stable function α as follows:

[0156]

[0157] Where c1 is a positive constant, and the derivative of the stable function is obtained:

[0158]

[0159] Substituting the stability function into In the equation, we get:

[0160]

[0161] When z2 is 0, then

[0162] Since x2 does not need to be constrained, V1 is augmented using a quadratic function and a Lyapunov function is designed as follows:

[0163] V3=V1+V2

[0164] in, Taking the derivative of the Lyapunov function V3, we get:

[0165]

[0166] Let F(X)=f(x,t)+β and Then we can get:

[0167]

[0168] In specific implementation, as a preferred embodiment of the present invention, step S3 specifically includes:

[0169] S31, F(X) is unknown, according to the function approximation theorem, using the fuzzy logic system θ *T ε(j) approximates the unfamiliar continuous function F(X). According to step S22, a fuzzy logic system is designed as follows:

[0170] F(X)=θ *T ε(j)+ω

[0171] in, represents the ideal weight, represents the fuzzy basis vector, ω represents the approximation error of the fuzzy logic system, and F(X) is a bounded function;

[0172] S32. Calculate the estimated value of the bounded function F(X). The calculation formula is as follows:

[0173]

[0174] in, is θ * An estimate of Make and

[0175] S33, Setting Note 1: Define time-varying gain n = 1 / q1(t) min , where q1(t) min It is the lower bound of q1(t). Since the value of q1(t) is unknown and time-varying, an adaptive method is used to estimate their boundaries and estimate the lower bound of q1(t) to avoid the singular problem when it is estimated to be zero.

[0176] In specific implementation, as a preferred embodiment of the present invention, step S4 specifically includes:

[0177] S41. Design the adaptive fuzzy non-quantized control input signal and adaptive law as follows:

[0178]

[0179] Among them, there is an adaptive law and γ1,γ2,p,η,k2,c2,c,λ1 are all positive constants greater than zero, represents an estimate of n, and

[0180] S42, control input signal And the state variables, intermediate signals, and adaptive laws in the system are quantified as follows:

[0181]

[0182] S43. Consider input quantization but not state quantization, and prove stability;

[0183] S44. Prove that the quantization error is bounded;

[0184] S45. Consider both input quantization and state quantization to prove stability.

[0185] In specific implementation, as a preferred embodiment of the present invention, step S43 specifically includes:

[0186] S431. Assume Lemma 3: Under the premise of considering the USV heading tracking control problem of Lemma 1 and Lemma 2, together with the control input, the adaptive law, the closed-loop control system error signal is consistent and ultimately bounded;

[0187] S432. Design the Lyapunov function as follows:

[0188]

[0189] The time derivative of the above equation is as follows:

[0190]

[0191] Substituting the control law into the above formula, we get:

[0192]

[0193] Substituting the adaptive law into the above formula, we get:

[0194]

[0195] S433、Set because so And because the time-varying gain n=1 / q1(t) min , so we introduce:

[0196]

[0197] Substituting into the above formula, we get:

[0198]

[0199] By the following formula:

[0200]

[0201]

[0202] -λ1z2sgnz2≤-λ1|z2|

[0203] Then it can be deduced that:

[0204]

[0205] According to Lemma 2, we get Then we get:

[0206]

[0207] S434, Order According to Lemma 1 and Lyapunov stability theory, the error of the USV heading tracking control system is bounded when state quantization is not considered, and the closed-loop system signal is ultimately bounded; and z1, z2, Bounded and obtainable It satisfies |z1|<k b ,t∈(0,∞); when t tends to infinity, V(t) converges to And its convergence accuracy depends on c. When c is much larger than λ, V(t)→0,z1→0,z2→0.

[0208] In specific implementation, as a preferred embodiment of the present invention, step S44 specifically includes:

[0209] S441. Set Lemma 4: Define the error surface, control input, state variable, and quantization error of the intermediate signal as

[0210] S442, from steps S22 and S42, β α 、 and Bounded, as follows:

[0211]

[0212] S443. From Lemma 3, we can see that the closed-loop system is stable when state quantization is not considered, and the error signal of the closed-loop control system is consistent and ultimately bounded. It is bounded.

[0213] S444: Launch according to step S41 According to step S22, it is deduced:

[0214]

[0215] And C and ν are both positive constants;

[0216] S445. According to step S41 and step S42, we have:

[0217]

[0218] S446: According to steps S41 and S42, β τ Bounded and combined with the formula in step S445, we can know that:

[0219]

[0220] Then there is a constant Make It can be seen from this that the quantization error in the designed quantitative feedback control system is bounded;

[0221] S447, Set Note 2: Because is a positive constant; z1, z2, is bounded, then we have is bounded, and o2 and δ are both bounded constants, where:

[0222]

[0223] S448. Considering the quantized state variables, intermediate signal formulas and control laws according to step S42, it is proved that the designed USV heading tracking control closed-loop system with quantization is stable, and the closed-loop system signals are consistent and ultimately bounded.

[0224] In specific implementation, as a preferred embodiment of the present invention, step S45 specifically includes:

[0225] S451. Design the time derivative of the Lyapunov function as follows:

[0226]

[0227] S452. According to step S44, the state quantization error of the designed quantization feedback system is bounded, and Then launch:

[0228]

[0229] S453, Order According to Lemma 1 and Lyapunov stability theory, it is proved that the USV heading tracking control closed-loop system with state quantization and input quantization is stable, and the closed-loop system signal is consistent and ultimately bounded.

[0230] Example

[0231] In order to verify the research method of the present invention for the adaptive fuzzy heading tracking control of unmanned vessels based on input quantization and state quantization input and output constraints, this embodiment uses MATLAB for computer simulation research. The parameters are set as follows:

[0232] The simulation object is the Dalian Maritime University USV "Lanxin". The USV parameters are: K = 0.71, T = 0.32, b = K / T = 2.2, a = 0.001. The controller parameters are: c1 = 20, c2 = 30, o1 = 0.1, γ1 = 0.1, γ2 = 1, η = 0.2, k2 = 5, λ1 = 20. The desired heading command is set to The initial state of the actual controlled object is [4, 2], and the quantization level χ = 0.1 is selected.

[0233] A uniform quantizer is used to process the state variables x1, x2 and the control input The control algorithm is written in MATLAB to simulate the closed-loop system under quantitative communication. The fuzzy system is designed to estimate the uncertainty F(X), using Gaussian membership function and a rule base size of 5×5. The boundedness of the quantization error is analyzed based on the Lyapunov function, and the convergence process of the tracking error is recorded. A desired heading and desired yaw rate are designed for simulation experiments to test whether the actual heading and actual yaw rate can successfully track the desired heading and desired yaw rate. The simulation results are calculated using the control algorithm written in MATLAB. Figure 2-Figure 5 shown.

[0234] exist Figure 2-Figure 5 The USV heading tracking results and tracking errors in Case 1, the fuzzy logic system estimation results, and the control inputs before and after quantization are shown in Figure 1. Specifically, Figure 2 The designed control strategy is given to enable the actual heading of the unmanned ship to accurately track the desired heading, indicating that the control strategy has good tracking performance. Figure 3 It is shown that under the interference of quantization error, external disturbance and system uncertainty, the tracking error of the designed controller gradually decreases over time and finally stabilizes in a smaller range. Figure 4 The approximation results of the fuzzy logic system for uncertainty terms and external disturbances are given. As can be seen from the figure, the estimation curve of the fuzzy logic system is close to the original function curve. This shows that the fuzzy logic system can well estimate the uncertainty terms and external disturbances in the system. Figure 5The control inputs before and after quantization are shown. When each 0.1 N·m change in the control input is considered a controller execution, the controller executed 4001 times without quantization and 2685 times after quantization. This clearly shows that considering state and input quantization effectively reduces the actuator's execution frequency, alleviating the communication burden and improving system resource efficiency while maintaining system performance.

[0235] The simulation results show that the introduction of state and input quantization and input-output constraints will not reduce the stability of the control system, which further proves the scientific nature of the designed control strategy.

[0236] 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. An adaptive fuzzy heading tracking control method for an unmanned ship with input quantization, state quantization and input-output constraints, characterized in that: include: S1. The uniform quantizer is used to quantize the input and state, and the quantization process is described linearly to illustrate the quantization process. S2. Introduce the logarithmic barrier Lyapunov function to ensure that the system state remains within a reasonable range; S3. Use fuzzy logic system to estimate uncertainty terms in the ship model and reduce the impact of external disturbances on the control system; S4. Based on Lyapunov stability theory and the proposed lemma, the boundedness of the quantization error and the stability of the designed control strategy are proved.

2. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 1 is characterized in that: Step S1 specifically includes: S11. Establish a mathematical model for USV heading control as follows: Among them, ψ represents the ship's heading angle; r represents the ship's yaw rate; β represents the disturbance caused by factors such as wind, waves, and currents; τ represents the control input of the system; b represents the control system gain, Where K represents the ship's turning index, T represents the ship's following index; a1 and a2 both represent the nonlinear coefficients of the Norrbin motion model. S12. Define the state variables as ψ and r, let ψ = x1, r = x2, and establish the mathematical model of quantized USV heading control as follows: Where Q(τ) is the quantized control input, f(x,t)=a1r+a2r 3 ,The quantization process is to convert a continuous signal into a discrete signal; S13. Assuming external disturbance and is a positive constant; S14. Design a uniform quantizer for the state variables x1, x2 and the control input Quantization is performed to obtain Q(g), as follows: in, In the above formula, χ represents the quantization step size, and χ>0, L i =χ,L i+1 =L i +χ, i∈R;, quantization error will be generated during the quantization process, and the quantization error satisfies S15. Let Q(τ) = q1(t)τ + q2(t), then: Among them, q1(t) is unknown and its sign remains unchanged during the quantization process. From the above formula, we can see that q1(t)>0. When |τ(t)|<a, Q(τ(t)) is also bounded. From this, we can infer that q2(t) is bounded.

3. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 1 is characterized in that: Step S2 specifically includes: S21. Introduce input constraints in the controller and use the saturation function sat(τ) instead of the sign function sgn(τ) as follows: S22. Introduce the logarithmic barrier Lyapunov function, as follows: Setting Lemma 1: For the error system where g = [z1z2] T , there exist continuously differentiable and positive definite functions V1 and V2; there exist two positive constants k a1 and k b1 , the position output is x1, and the heading signal error z1 is defined as x1-x 1d , when -k is satisfied a1 <z1<k b1 When V1(z1)→∞, γ1(||z2||)≤V2(z2)≤γ2(||z2||), γ1 and γ2 are K ∞ Class function; when -k a1 <z1(0)<k b1 , take V(z)=V1(z1)+V2(z2), if but z1(t)∈(-k a1 ,k b1 ), where λ and μ are constants; Lemma 2: For k b1 >0, if full Then there is On the error surface z1=x1-x 1d , z2=x2-α, α is a stable function that has not yet been designed. In order to achieve Define the logarithmic barrier function as follows: Where log(·) is the natural logarithm, k b1 =k d -B1 represents the constraint on z1, and we get |z1|<k b1 , and it can also be concluded that V1 is in |z1|<k b1 is positive definite, and taking the derivative of V1, we get: Design a stable function α as follows: Where c1 is a positive constant, and the derivative of the stable function is obtained: Substituting the stability function into In the equation, we get: When z2 is 0, then Since x2 does not need to be constrained, V1 is augmented using a quadratic function and a Lyapunov function is designed as follows: V3=V1+V2 in, Taking the derivative of the Lyapunov function V3, we get: Let F(X)=f(x,t)+β and Then we can get:

4. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 1 is characterized in that: Step S3 specifically includes: S31, F(X) is unknown, according to the function approximation theorem, using the fuzzy logic system θ *T ε(j) approximates the unfamiliar continuous function F(X). According to step S22, a fuzzy logic system is designed as follows: F(X)=θ *T e(j)+ω in, represents the ideal weight, represents the fuzzy basis vector, ω represents the approximation error of the fuzzy logic system, and F(X) is a bounded function; S32. Calculate the estimated value of the bounded function F(X). The calculation formula is as follows: in, is θ * An estimate of Make and S33, Setting Note 1: Define time-varying gain n = 1 / q1(t) min , where q1(t) min It is the lower bound of q1(t). Since the value of q1(t) is unknown and time-varying, an adaptive method is used to estimate their boundaries and estimate the lower bound of q1(t) to avoid the singular problem when it is estimated to be zero.

5. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 1, characterized in that: Step S4 specifically includes: S41. Design the adaptive fuzzy non-quantized control input signal and adaptive law as follows: Among them, there is an adaptive law and γ1,γ2,p,η,k2,c2,c,λ1 are all positive constants greater than zero, represents an estimate of n, and S42, control input signal And the state variables, intermediate signals, and adaptive laws in the system are quantified as follows: S43. Consider input quantization but not state quantization, and prove stability; S44. Prove that the quantization error is bounded; S45. Consider both input quantization and state quantization to prove stability.

6. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 5, characterized in that: Step S43 specifically includes: S431. Assume Lemma 3: Under the premise of considering the USV heading tracking control problem of Lemma 1 and Lemma 2, together with the control input, the adaptive law, the closed-loop control system error signal is consistent and ultimately bounded; S432. Design the Lyapunov function as follows: The time derivative of the above equation is as follows: Substituting the control law into the above formula, we get: Substituting the adaptive law into the above formula, we get: S433、Set because so And because the time-varying gain n=1 / q1(t) min , so we introduce: Substituting into the above formula, we get: By the following formula: Then it can be deduced that: According to Lemma 2, we get Then we get: S434, Order According to Lemma 1 and Lyapunov stability theory, the error of the USV heading tracking control system is bounded when state quantization is not considered, and the closed-loop system signal is ultimately bounded; and z1, z2, Bounded and obtainable It satisfies |z1|<k b ,t∈(0,∞); when t tends to infinity, V(t) converges to And its convergence accuracy depends on c. When c is much larger than λ, V(t)→0,z1→0,z2→0.

7. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 5, characterized in that: Step S44 specifically includes: S441. Set Lemma 4: Define the error surface, control input, state variable, and quantization error of the intermediate signal as S442, from steps S22 and S42, and Bounded, as follows: S443. From Lemma 3, we can see that the closed-loop system is stable when state quantization is not considered, and the error signal of the closed-loop control system is consistent and ultimately bounded. It is bounded. S444: Launch according to step S41 According to step S22, it is deduced: And C and ν are both positive constants; S445. According to step S41 and step S42, we have: S446: According to steps S41 and S42, β τ Bounded and combined with the formula in step S445, we can know that: Then there is a constant Make It can be seen from this that the quantization error in the designed quantitative feedback control system is bounded; S447, Set Note 2: Because is a positive constant; z1, z2, is bounded, then we have is bounded, and o2 and δ are both bounded constants, where: S448. Considering the quantized state variables, intermediate signal formulas and control laws according to step S42, it is proved that the designed USV heading tracking control closed-loop system with quantization is stable, and the closed-loop system signals are consistent and ultimately bounded.

8. The method for adaptive fuzzy heading tracking control of an unmanned vessel with input quantization, state quantization and input-output constraints according to claim 5, characterized in that: Step S45 specifically includes: S451. Design the time derivative of the Lyapunov function as follows: S452. According to step S44, the state quantization error of the designed quantization feedback system is bounded, and Then launch: S453, Order According to Lemma 1 and Lyapunov stability theory, it is proved that the USV heading tracking control closed-loop system with state quantization and input quantization is stable, and the closed-loop system signal is consistent and ultimately bounded.

Images