Unmanned ship dynamic event triggering obstacle avoidance control strategy based on actuator faults

Abstract

The invention provides an unmanned ship dynamic event trigger obstacle avoidance control strategy based on an actuator fault, and the strategy comprises the steps: carrying out the stress analysis of an unmanned ship, and building an unmanned ship dynamic model under the actuator fault through combining with an actuator fault model; constructing an improved artificial potential field function according to the information of the unmanned ship and the obstacle information; for lumped disturbance composed of actual control input, internal model uncertainty and external disturbance, an extended state observer is adopted for estimation; and based on the extended state observer and the constructed dynamic event triggering mechanism, designing an unmanned ship dynamic event triggering obstacle avoidance control law by adopting a backstepping method, and realizing control of unmanned ship dynamic event triggering obstacle avoidance under the condition of an actuator fault. According to the unmanned ship dynamic event triggering obstacle avoidance control strategy based on the actuator fault, the obstacle avoidance performance of the unmanned ship can be guaranteed, the tracking performance of the unmanned ship can be guaranteed, and the updating frequency of the actuator can be reduced by introducing a dynamic event triggering mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned vessel control, and in particular to an unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure. Background Art

[0002] In recent years, the control of unmanned vessels (UAVs) has been extensively studied. As common autonomous vehicles, UAVs are widely used to perform dangerous and repetitive tasks due to their miniaturization, intelligence, and versatile marine transport platforms. During navigation, UAVs are subject to interference from internal model uncertainty and external disturbances, resulting in significant tracking errors. Compensating for these disturbances is crucial for improving UAV tracking accuracy. Furthermore, actuator failures are inevitable due to factors such as actuator aging and seawater corrosion. Regardless of the nature of these failures, they can cause significant performance loss or even mission cancellation. Although researchers have proposed a general fault-tolerant control strategy for actuator failure models, these strategies are designed for linearized ship models. The real-world marine environment is highly complex, with numerous static and dynamic obstacles such as reefs, shipwrecks, and currently operating vessels, posing significant challenges to the safe navigation of UAVs. To reduce their dependence on shore operators, UAVs should possess autonomous decision-making and intelligent obstacle avoidance capabilities. They should utilize external sensors to sense environmental information and autonomously avoid obstacles in real time. Numerous obstacle avoidance control methods have been developed, including speed barrier methods, artificial potential field methods, reinforcement learning-based collision avoidance algorithms, and vector field-based obstacle avoidance algorithms. Artificial potential field functions have been widely studied due to their simple structure and strong real-time performance. However, actuator failures can render existing obstacle avoidance control strategies ineffective and potentially lead to collisions. Furthermore, continuous actuator updates cause wear and tear. Reducing the actuator update frequency is crucial for extending actuator life and minimizing the risk of failure. Summary of the Invention

[0003] In order to solve the technical problems in the prior art that actuator failure may render the existing obstacle avoidance control strategy ineffective, possibly leading to collision, and that continuous updates of the actuator may cause actuator wear, thereby reducing the update frequency of the actuator and extending the service life of the actuator, the technical means adopted by the present invention are as follows: an unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure, comprising the following steps:

[0004] Conduct force analysis on the unmanned vessel and, combined with the actuator failure model, establish a dynamic model of the unmanned vessel under actuator failure.

[0005] According to the information of the unmanned ship itself and the obstacle information, an improved artificial potential field function is constructed;

[0006] An extended state observer is established to estimate the lumped disturbance of unknown actual control input, internal model uncertainty of the UAV and external disturbance.

[0007] Based on the extended state observer and the constructed dynamic event triggering mechanism, the backstepping method is used to design the unmanned ship dynamic event triggered obstacle avoidance control law, which can realize the dynamic event triggered obstacle avoidance control of the unmanned ship in the event of actuator failure.

[0008] Furthermore, the specific process of performing force analysis on the unmanned ship and establishing the unmanned ship dynamics model under actuator failure in combination with the actuator failure model is as follows:

[0009] S11. Define the desired parameterized path of the unmanned ship as follows:

[0010]

[0011] Where: θ represents the path variable, [x d (θ),y d (θ)] T represents the desired position of the unmanned ship, ψ d (θ) represents the desired heading angle of the unmanned ship;

[0012] S12. Define the kinematic model of the unmanned ship as follows:

[0013]

[0014] where J(ψ) represents the rotation matrix, ν(t) = [u, v, r] T Indicates the speed of the unmanned ship;

[0015] S13. Based on the actuator failure model of the unmanned ship, a dynamic model of the unmanned ship under actuator failure is established;

[0016] The specific fault model of the unmanned ship actuator is:

[0017] τ F (t)=Ψτ(t)+Πτ s (t)

[0018] Among them, τ F is the unknown actual control input, τ(t) is the control input, τ s (t) is the unknown function caused by bias fault or stuck fault, Ψ=diag{Ψ1,...,Ψ i ,...,Ψ m}、Π=diag{Π1,...,Π i ,...,Π m} is the unknown positive semidefinite weight matrix,

[0019] The force analysis of the unmanned boat is carried out to obtain the dynamic model of the unmanned boat:

[0020]

[0021] Where M is the inertia matrix of the unmanned ship; C(ν) is the Coriolis force centripetal force matrix; D(ν) is the damping matrix; g(ν,η) represents the unmodeled fluid mechanics; τ w (t) represents the external ocean environment disturbance caused by wind, waves, and currents; G represents the actuator allocation matrix.

[0022] Furthermore, the specific process of constructing the improved artificial potential field function based on the unmanned ship's own information and obstacle information is as follows:

[0023] S21. In order to compensate for the impact of actuator failure, an improved artificial potential field function is designed:

[0024]

[0025] make And taking partial derivative of η we get:

[0026]

[0027] Where: p rk =η-p k ,η represents the position of the unmanned ship, p k represents the position of the kth obstacle, Θ is a positive adjustable parameter, is the maximum detection distance, represents the lower bound of the actuator failure, so in the interval middle, It shows a monotonically decreasing trend. When ||p rk ||Close d hour, tends to infinity, and furthermore,

[0028] Furthermore, the extended state observer is established as follows:

[0029] S31. In order to estimate the lumped disturbance consisting of the actual control input, internal model uncertainty and external disturbance, the following extended state observer is used:

[0030]

[0031] in, and denote the estimated values ​​of ν(t) and σ(t) respectively; σ(t)=M -1 [-C(ν)ν(t)-D(ν)ν(t)-g(ν,η)+τw (t)]+M -1 Hτ F -M -1 Hτ represents the lumped disturbance,

[0032] In addition, assuming that σ(t) is bounded, that is, Extended state observer matrix K1 = 2wI3, K2 = w 2 I3, and w represents the observer bandwidth.

[0033] Furthermore, the dynamic event triggering mechanism based on the extended state observer and the constructed dynamic event triggering mechanism adopts the backstepping method to design the unmanned ship dynamic event triggered obstacle avoidance control law, and realizes the control process of the unmanned ship dynamic event triggered obstacle avoidance in the case of actuator failure. The specific process is as follows:

[0034] S41. Assume τ s (t) is bounded, that is, ||τ s (t)||≤Λ, where Λ is a positive constant, then there exists a positive number Make For all All are established; is the actuator failure lower bound factor;

[0035] Define tracking error e η :

[0036] e η =J T (η-η d (θ0)

[0037] For e η Taking the derivative we get:

[0038]

[0039] S42, unmanned ship kinematic controller ν r The design is as follows:

[0040]

[0041] The path update law of the unmanned boat is designed as follows:

[0042]

[0043] Where k1=diag{k 11 ,k 12 ,k 13}; l and μ are positive numbers; z = e η +J T z3 and

[0044]

[0045] S43. Definition in is ν r The estimated value of By ν r After a first-order filter, we can get

[0046]

[0047] Where: t d is a time constant; so that the derivative of the kinematic controller b j is bounded and satisfies is a positive constant;

[0048] S44, dynamic event triggered obstacle avoidance controller design is as follows:

[0049]

[0050] Where, the controller gain k2=diag{k 21 ,k 22 ,k 23}, f(t) is a bounded continuous function. In order to avoid high gain, f(t) is designed as follows:

[0051] f(t)=e -st

[0052] Where s is a positive number;

[0053] It is derived from the following adaptive law:

[0054]

[0055] Where, Ω is a given positive number;

[0056] S45. The dynamic event triggering mechanism is designed as follows:

[0057]

[0058] and

[0059]

[0060] where i=1,2,...,m,a i Satisfying 0<a i <a * , where a * is a positive constant, Indicates the update time, k is a number of trigger events, is a dynamic variable, defined as follows:

[0061]

[0062] Where, is a positive number; Considered to be The filtered value of

[0063] According to the dynamic event triggering mechanism, there is a constant μ i Satisfy |μ i |≤1 makes Right now

[0064]

[0065] where μ=diag{μ1,μ2,...,μ m}; a=[a1,a2,...,a m ] T ;

[0066] Furthermore, it also includes selecting a suitable Lyapunov function based on the extended state observer and proving that the error subsystem of the extended state observer is input-to-state stable. The specific process is as follows:

[0067] S51. Define the Lyapunov energy function V1 as follows:

[0068] V1=(1 / 2)E T (t)QE(t)

[0069] Taking the derivative of V1, we get:

[0070]

[0071] when When , we get:

[0072]

[0073] It follows that the observer error subsystem is input-state stable; note that V1 is bounded and satisfies ([λ min (Q)] / 2)||E(t)|| 2 ≤V1≤([λ max (Q)] / 2)||E(t)|| 2 ,get:

[0074]

[0075] in, 0<κ<1, κ is a constant, and furthermore, assuming is established, in which It is a positive number.

[0076] Furthermore, it also includes the unmanned ship dynamic event triggered obstacle avoidance control law, selects the appropriate Lyapunov function, and proves that the dynamic event triggered obstacle avoidance control subsystem is input to the state stability; the specific process is as follows:

[0077] S61. Define the Lyapunov energy function as follows:

[0078]

[0079] Taking the derivative of V2, we get:

[0080]

[0081] in: choose t d is the time constant;

[0082] Case 1: When t→∞, there is a continuous function f=e -st →0, and outside the obstacle avoidance range have Therefore, z = e η , then V2 becomes:

[0083]

[0084] Where, Error Matrix

[0085] From this we can get: when hour,

[0086] Case 2: Outside the obstacle avoidance range, i.e. When t→∞, f=e -st →0, we get:

[0087]

[0088] Where, From this we can get: when hour,

[0089] So we can get that the control error subsystem is input to the state stable.

[0090] Furthermore, the stability analysis method of the cascade system is applied to prove that the closed-loop system is input-state stable and does not suffer from the Zeno phenomenon. The specific process is as follows:

[0091] Based on the stability condition of the cascade system, it can be obtained that the closed-loop system composed of the extended state observer and the dynamic event-triggered obstacle avoidance controller is input-state stable;

[0092] The proposed dynamic event triggering mechanism avoids Zeno behavior when in Established;

[0093] According to the dynamic event triggering mechanism, we get for Learn is bounded, so there exists a positive constant W i * Make Considering e i (t k )=0 and Available

[0094] In conclusion, the proposed dynamic event triggering mechanism avoids the occurrence of Zeno phenomenon.

[0095] This application proposes a dynamic event-triggered obstacle avoidance control strategy for an unmanned vessel based on actuator failure. By considering the impact of actuator failure, an improved artificial potential field function is constructed. Furthermore, considering the internal model uncertainty, external disturbances, and other interferences of the unmanned vessel, the lumped disturbance consisting of the actual control input, internal model uncertainty, and external disturbances is estimated. Finally, a dynamic event-triggered obstacle avoidance control strategy for an unmanned vessel based on actuator failure is proposed using the backstepping method. Compared to traditional obstacle avoidance control strategies, the control strategy proposed in this invention can not only compensate for the impact of actuator failures and reduce the actuator update frequency, but also maintain good obstacle avoidance performance.

[0096] Compared with the prior art, the present invention has the following advantages:

[0097] 1. The proposed actuator failure-based dynamic event-triggered obstacle avoidance control strategy for the unmanned vehicle not only ensures the unmanned vehicle's obstacle avoidance performance but also its tracking performance. Furthermore, compared to traditional control strategies, the introduction of a dynamic event-triggered mechanism reduces the actuator update frequency and reduces the burden on the actuator.

[0098] 2. This invention addresses the situation of actuator failures, taking into account the uncertainty of model parameters and the impact of external disturbances, and establishes an extended state observer with actuator failures. Furthermore, an improved artificial potential field function is designed to account for the impact of actuator failures. Based on this, a dynamic event-triggered obstacle avoidance control strategy is designed. This strategy reduces actuator wear while ensuring system stability, thereby reducing the likelihood of actuator failure and improving the system's fault tolerance.

[0099] 3. The present invention takes into account the occurrence of actuator failure and considers the failure information in the dynamic event triggering mechanism, so that the stability of the system can be guaranteed when the actuator failure occurs.

[0100] Based on the above reasons, the present invention can be widely promoted in fields such as unmanned ships. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] 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.

[0102] Figure 1 Flow chart of the method of the present invention.

[0103] Figure 2 This is a simulation diagram of the control effect provided by an embodiment of the present invention.

[0104] Figure 3 This is a speed simulation diagram provided by an embodiment of the present invention.

[0105] Figure 4 This is a simulation diagram of the path tracking error provided by an embodiment of the present invention.

[0106] Figure 5 This is a control input simulation diagram provided by an embodiment of the present invention.

[0107] Figure 6 This is a simulation diagram of the number of actuator updates provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0108] 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.

[0109] like Figure 1 As shown, the present invention provides an unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure, comprising the following steps:

[0110] S1. Conduct force analysis on the unmanned vessel and, combined with the actuator failure model, establish a dynamic model of the unmanned vessel under actuator failure.

[0111] S2. Constructing an improved artificial potential field function based on the UAV's own information (i.e., position information and speed information, etc.) and obstacle information (including obstacle position information, etc.);

[0112] S3, the extended state observer is used to estimate the unknown actual control input, the internal model uncertainty of the unmanned ship and the lumped disturbance of the external disturbance;

[0113] The internal model uncertainty refers to the internal model uncertainty caused by the fact that the model parameters are related to the velocity, are time-varying, and contain unknown fluid mechanics effects, that is,

[0114] C(ν)ν(t)+D(ν)ν(t)+g(ν,η)

[0115] S4. Based on the extended state observer and the constructed dynamic event triggering mechanism, the backstepping method is used to design the unmanned ship dynamic event triggered obstacle avoidance control law to realize the control of unmanned ship dynamic event triggered obstacle avoidance in the event of actuator failure.

[0116] After step S1 is executed, step S2 and step S3 are executed sequentially, and then step S4 is executed;

[0117] Furthermore, based on the established extended state observer, a suitable Lyapunov function is selected, and it is proved that the error subsystem of the extended state observer is input-to-state stable.

[0118] Based on the designed control law, a suitable Lyapunov function is selected to prove that the dynamic event triggered obstacle avoidance control subsystem is input to state stable.

[0119] By applying the stability analysis method of cascade system, it is proved that the closed-loop system is input-state stable and there is no Zeno phenomenon.

[0120] The cascade system includes an extended state observer and an actuator;

[0121] The specific process of step S1: performing force analysis on the unmanned vessel and establishing the unmanned vessel dynamics model under actuator failure in combination with the actuator failure model is as follows:

[0122] S11. Define the parameterized path expected by the unmanned ship as follows:

[0123]

[0124] Where: θ represents the path variable, [x d (θ),y d (θ)] T represents the desired position of the unmanned ship, ψ d (θ) represents the desired heading angle of the unmanned ship.

[0125] S12. Define the kinematic model of the unmanned ship as follows:

[0126]

[0127] Where: J(ψ) represents the rotation matrix, ν(t)=[u,v,r] T Indicates the speed of the unmanned ship;

[0128] S13. Based on the actuator failure model, a dynamic model of the unmanned ship under actuator failure is established;

[0129] The actuator fault model is specifically:

[0130] τ F (t)=Ψτ(t)+Πτ s (t)

[0131] Among them, τ F is the unknown actual control input, τ(t) is the control input, τ s (t) is the unknown function caused by bias fault or stuck fault. Ψ=diag{Ψ1,...,Ψ i ,...,Ψ m}、Π=diag{Π1,...,Π i ,...,Π m} is the unknown positive semidefinite weight matrix.

[0132] Perform force analysis on the unmanned boat and obtain its dynamic model:

[0133]

[0134] Where M is the inertia matrix of the unmanned ship; C(ν) is the Coriolis force centripetal force matrix; D(ν) is the damping matrix; g(ν,η) represents the unmodeled fluid mechanics; τ w (t) represents the external ocean environment disturbance caused by wind, waves, and currents; G represents the actuator allocation matrix.

[0135] Furthermore, based on the UAV’s own information and obstacle information, the specific process of constructing the improved artificial potential field function is as follows:

[0136] S21. In order to compensate for the impact of actuator failure, an improved artificial potential field function is designed:

[0137]

[0138] make And taking partial derivative of p we get:

[0139]

[0140] Where: p rk =η-p k η represents the position of the unmanned ship, p k represents the position of the kth obstacle. Θ is a positive adjustable parameter. is the maximum detection distance. represents the lower bound of the actuator failure. Therefore, in the interval middle, It shows a monotonically decreasing trend. When ||p rk ||Close d hour, tends to infinity. In addition,

[0141] Furthermore, the specific process of establishing an extended state observer to estimate the lumped disturbance of unknown actual control input, internal model uncertainty of the unmanned ship, and external disturbance is as follows:

[0142] S31. In order to estimate the lumped disturbance consisting of the actual control input, the internal model uncertainty of the unmanned ship and the external disturbance, the following extended state observer is used:

[0143]

[0144] in, and denote the estimated values ​​of ν(t) and σ(t) respectively; σ(t)=M -1 [-C(ν)ν(t)-D(ν)ν(t)-g(ν,η)+τ w (t)]+M -1 Hτ F -M -1 Hτ represents the lumped disturbance, Furthermore, it is natural to assume that σ(t) is bounded, i.e. Extended state observer matrix K1 = 2wI3, K2 = w 2 I3, and w represents the observer bandwidth.

[0145] Furthermore, based on the extended state observer and the constructed dynamic event triggering mechanism, the backstepping method is used to design the unmanned ship dynamic event triggered obstacle avoidance control law. The specific process of realizing the control of the unmanned ship dynamic event triggered obstacle avoidance in the case of actuator failure is as follows:

[0146] S41. Assume τ s (t) is bounded, that is, ||τ s (t)||≤Λ. Where Λ is a positive constant. Then, there exists a positive number Make For all All are established. is the actuator failure lower bound factor;

[0147] Define tracking error e η :

[0148] e η =J T (η-η d (θ0)

[0149] For e η Taking the derivative we get

[0150]

[0151] S42, the kinematic controller is designed as follows:

[0152]

[0153] And the path update law is designed as follows:

[0154]

[0155] Where k1=diag{k 11 ,k 12 ,k 13}; l and μ are positive numbers; z = e η +J T z3 and

[0156]

[0157] S43. Definition in is ν r estimated value. By ν r After a first-order filter, we can get

[0158]

[0159] where t d is the time constant. b j is bounded and satisfies is a positive constant.

[0160] S44, dynamic event triggered obstacle avoidance control law is designed as follows:

[0161]

[0162] Where, k2=diag{k 21 ,k 22 ,k 23}. f(t) is a bounded continuous function. In order to avoid high gain, f(t) is designed as follows:

[0163] f(t)=e -st

[0164] Where s is a positive number.

[0165] It is derived from the following adaptive law:

[0166]

[0167] Here, Ω is a given positive number.

[0168] S45. The dynamic event triggering mechanism is designed as follows:

[0169]

[0170] and

[0171]

[0172] where i=1,2,...,m. a i Satisfying 0<a i <a * , where a * is a positive constant. represents the update time, and k is the number of triggering events. is a dynamic variable, defined as follows:

[0173]

[0174] Where, is a positive number; Can be regarded as The filter value of .

[0175] According to the dynamic event triggering mechanism, there exists μ i Satisfy |μ i |≤1 makes Right now

[0176]

[0177] where μ=diag{μ1,μ2,...,μ m}; a=[a1,a2,...,a m ] T ;

[0178] Furthermore, based on the extended state observer, we select a suitable Lyapunov function and prove that the error subsystem of the extended state observer is stable in terms of input to state. The specific process is as follows:

[0179] S51. Define the Lyapunov energy function as follows:

[0180] V1=(1 / 2)E T (t)QE(t)

[0181] Taking the derivative of V1, we get:

[0182]

[0183] when , we can get:

[0184]

[0185] It follows that the observer error subsystem is input-state stable. Note that V1 is bounded and satisfies ([λ min (Q)] / 2)||E(t)|| 2 ≤V1≤([λ max (Q)] / 2)||E(t)|| 2 . We can get:

[0186]

[0187] in, 0<κ<1, κ is a constant. In addition, assume is established, in which It is a positive number.

[0188] Furthermore, based on the unmanned ship dynamic event-triggered obstacle avoidance control law, a suitable Lyapunov function is selected to prove that the dynamic event-triggered obstacle avoidance control subsystem is input to the state stability. The specific process is as follows:

[0189] S61. Define the Lyapunov energy function as follows:

[0190]

[0191] Taking the derivative of V2, we get:

[0192]

[0193] in: choose

[0194] Case 1: When t→∞, f=e -st →0, and outside the obstacle avoidance range have Therefore, z = e η Then V2 can be transformed into:

[0195]

[0196] Where, From this we can get: when hour,

[0197] Case 2: Outside the obstacle avoidance range, i.e. When t→∞, f=e -st → 0. We can get:

[0198]

[0199] Where, From this we can get: when hour, No specific meaning.

[0200] So we can get that the control error subsystem is input to the state stable.

[0201] Furthermore, by applying the stability analysis method of cascade systems, it is proved that the closed-loop system is input-state stable and does not suffer from the Zeno phenomenon. The specific process is as follows:

[0202] S71. Based on the stability condition of the cascade system, it can be obtained that the closed-loop system composed of the extended state observer and the dynamic event-triggered controller is input-state stable.

[0203] Next, we will prove that the dynamic event triggering mechanism proposed in S4 can avoid Zeno behavior. In other words, where t i * > 0 is established. According to the dynamic event triggering mechanism, we can get for It can be known is bounded. Therefore, there exists a positive constant W i * Make Considering e i (t k )=0 and Available In summary, the dynamic event triggering mechanism proposed in S4 avoids the occurrence of Zeno phenomenon.

[0204] A simulation verification study was conducted on the unmanned ship dynamic model, dynamic event-triggered obstacle avoidance controller and adaptive update rate under the unmanned ship dynamic obstacle avoidance control strategy under actuator failure, and compared with conventional means to further verify the effectiveness and superiority.

[0205] In order to verify the effectiveness of the unmanned ship dynamic event triggered obstacle avoidance control strategy based on actuator failure provided in this embodiment, a simulation experiment was carried out using MATLAB and a detailed explanation was given.

[0206] The unmanned ship model adopted in this embodiment comprehensively considers actuator failures and external interference, adopts the backstepping method and the improved artificial potential field function, and designs an unmanned ship dynamic event-triggered obstacle avoidance controller under actuator failure. The controller can make the closed-loop system stable in input state, have good tracking performance and obstacle avoidance performance, have certain robustness against actuator failures, and have good suppression of external interference.

[0207] Specifically, in this embodiment, it is assumed that the reference path is

[0208] x d (θ0)=y d (θ0)=0.09θ0+3

[0209] ψ d =π / 4

[0210] The actuator allocation matrix of the unmanned ship model is:

[0211]

[0212] In the simulation, the observer bandwidth is set to w = 40, the observer parameters are k1 = diag{2,2,2}, k2 = diag{60,16.9,70}. The position of obstacle No. 1 is p1 = [1.5 3 0] T , the position of obstacle 2 is p2 = [4.5 4 0] T The controller parameters are set to: f(t) = 0.5e -0.01t ,Ω=10,u s =2,t d =0.1, The path update law parameters are set to: l = 5, μ = 5. The external disturbance is set to τ w =[4cos(2.5t)sin(0.5t) 3sin(0.5t)cos(t) 0.5sin(t)] T , the maximum detection range is The minimum safe distance is d= 1. Assume that an actuator failure occurs at 30 seconds and the fault parameters are Ψ = dia{0.8, 0, 0.5, 0.6, 0.8}.

[0213] Based on the above parameters, the proposed unmanned ship dynamic event triggered obstacle avoidance control method based on actuator failure is simulated and verified. Figure 2-6 As shown. Among them, Figure 2 Displaying the control effect diagram of the unmanned boat can ensure good obstacle avoidance and tracking performance; Figure 3 , displays the speed information of the unmanned ship; Figure 4 Display the tracking error of the unmanned vessel; Figure 5 Display control inputs for the unmanned vessel; Figure 6 The update times of each actuator are displayed. So far, the digital simulation of the algorithm has been completed and its effectiveness has been verified.

[0214] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0215] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0216] 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 dynamic event-triggered obstacle avoidance control strategy for an unmanned vessel based on actuator failure, characterized in that: The steps include: Conduct force analysis on the unmanned vessel and, combined with the actuator failure model, establish a dynamic model of the unmanned vessel under actuator failure. According to the information of the unmanned ship itself and the obstacle information, an improved artificial potential field function is constructed; An extended state observer is established to estimate the lumped disturbance of unknown actual control input, internal model uncertainty of the UAV and external disturbance. Based on the extended state observer and the constructed dynamic event triggering mechanism, the backstepping method is used to design the unmanned ship dynamic event triggered obstacle avoidance control law, which can realize the dynamic event triggered obstacle avoidance control of the unmanned ship in the event of actuator failure.

2. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1 is characterized in that: The specific process of conducting force analysis on the unmanned ship and establishing the unmanned ship dynamics model under actuator failure in combination with the actuator failure model is as follows: S11. Define the desired parameterized path of the unmanned ship as follows: in: Represents a path variable, represents the desired position of the unmanned vessel, represents the desired heading angle of the unmanned ship; S12. Define the kinematic model of the unmanned ship as follows: where J(ψ) represents the rotation matrix, ν(t) = [u, v, r] T Indicates the speed of the unmanned ship; S13. Based on the actuator failure model of the unmanned ship, a dynamic model of the unmanned ship under actuator failure is established; The specific fault model of the unmanned ship actuator is: t F (t)=Ψτ(t)+Πτ s (t) Among them, τ F is the unknown actual control input, τ(t) is the control input, τ s (t) is the unknown function caused by bias fault or stuck fault, Ψ=diag{Ψ1,...,Ψ i ,...,Ψ m }、Π=diag{Π1,...,Π i ,...,Π m } is the unknown positive semidefinite weight matrix, The force analysis of the unmanned boat is carried out to obtain the dynamic model of the unmanned boat: Where M is the inertia matrix of the unmanned ship; C(ν) is the Coriolis force centripetal force matrix; D(ν) is the damping matrix; g(ν,η) represents the unmodeled fluid mechanics; τ w (t) represents the external ocean environment disturbance caused by wind, waves, and currents; G represents the actuator allocation matrix.

3. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: The specific process of constructing the improved artificial potential field function based on the unmanned ship's own information and obstacle information is as follows: S21. In order to compensate for the impact of actuator failure, an improved artificial potential field function is designed: make And taking partial derivative of η we get: Where: p rk =η-p k ,η represents the position of the unmanned ship, p k represents the position of the kth obstacle, Θ is a positive adjustable parameter, is the maximum detection distance, represents the lower bound of the actuator failure, so in the interval middle, It shows a monotonically decreasing trend. When ||p rk ||Close d hour, tends to infinity, and furthermore, 4. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: The extended state observer is established as follows: S31. In order to estimate the lumped disturbance consisting of the actual control input, internal model uncertainty and external disturbance, the following extended state observer is used: in, and denote the estimated values ​​of ν(t) and σ(t) respectively; σ(t)=M -1 [-C(ν)ν(t)-D(ν)ν(t)-g(ν,η)+τ w (t)]+M -1 Hτ F -M -1 Hτ represents the lumped disturbance, In addition, assuming that σ(t) is bounded, that is, Extended state observer matrix K1 = 2wI3, K2 = w 2 I3, and w represents the observer bandwidth.

5. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: Based on the extended state observer and the constructed dynamic event triggering mechanism, the backstepping method is used to design the unmanned ship dynamic event triggered obstacle avoidance control law to achieve the control of the unmanned ship dynamic event triggered obstacle avoidance in the event of an actuator failure. The specific process is as follows: S41. Assume τ s (t) is bounded, that is, ||τ s (t)||≤Λ, where Λ is a positive constant, then there exists a positive number Make For all n=1,2,...,L are all valid; is the actuator failure lower bound factor; Define tracking error e η : For e η Taking the derivative we get: S42, unmanned ship kinematic controller ν r The design is as follows: The path update law of the unmanned boat is designed as follows: Where k1=diag{k 11 ,k 12 ,k 13 }; l and μ are positive numbers; z = e η +J T z3 and S43. Definition in is ν r The estimated value of By ν r After a first-order filter, we can get Where: t d is a time constant; so that the derivative of the kinematic controller b j is bounded and satisfies j=1,2,3, is a positive constant; S44, dynamic event triggered obstacle avoidance controller design is as follows: Where, the controller gain k2=diag{k 21 ,k 22 ,k 23 }, f(t) is a bounded continuous function. In order to avoid high gain, f(t) is designed as follows: f(t)=e -st Where s is a positive number; It is derived from the following adaptive law: Where, Ω is a given positive number; S45. The dynamic event triggering mechanism is designed as follows: and where i=1,2,...,m,a i Satisfying 0<a i <a * , where a * is a positive constant, Indicates the update time, k is a number of trigger events, is a dynamic variable, defined as follows: Where, is a positive number; Considered to be The filtered value of According to the dynamic event triggering mechanism, there is a constant μ i Satisfy |μ i |≤1 makes Right now Where μ = diag{μ1,μ2,...,μ m };a=[a1,a2,...,a m ] T ; 6. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: It also includes selecting a suitable Lyapunov function based on the extended state observer and proving that the error subsystem of the extended state observer is stable from input to state. The specific process is as follows: S51. Define the Lyapunov energy function V1 as follows: V1=(1 / 2)E T (t)QE(t) Taking the derivative of V1, we get: when When , we get: It follows that the observer error subsystem is input-state stable; note that V1 is bounded and satisfies ([λ min (Q)] / 2)||E(t)|| 2 ≤V1≤([λ max (Q)] / 2)||E(t)|| 2 ,get: in, 0<κ<1, κ is a constant, and furthermore, assuming is established, in which It is a positive number.

7. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: It also includes the unmanned ship dynamic event triggered obstacle avoidance control law, selects the appropriate Lyapunov function, and proves that the dynamic event triggered obstacle avoidance control subsystem is input to the state stability; the specific process is as follows: S61. Define the Lyapunov energy function as follows: Taking the derivative of V2, we get: in: choose t d is the time constant; Case 1: When t→∞, there is a continuous function f=e -st →0, and outside the obstacle avoidance range have Therefore, z = e η , then V2 becomes: Where, Error Matrix From this we can get: when hour, Case 2: Outside the obstacle avoidance range, i.e. When t→∞, f=e -st →0, we get: Where, From this we can get: when hour, So we can get that the control error subsystem is input to the state stable.

8. The unmanned vessel dynamic event-triggered obstacle avoidance control strategy based on actuator failure according to claim 1, characterized in that: It also includes the application of the stability analysis method of the cascade system, proving that the closed-loop system is input-state stable and there is no Zeno phenomenon. The specific process is as follows: Based on the stability condition of the cascade system, it can be obtained that the closed-loop system composed of the extended state observer and the dynamic event-triggered obstacle avoidance controller is input-state stable; The proposed dynamic event triggering mechanism avoids Zeno behavior when in Established; According to the dynamic event triggering mechanism, we get for Learn is bounded, so there exists a positive constant Make Considering e i (t k )=0 and Available In conclusion, the proposed dynamic event triggering mechanism avoids the occurrence of Zeno phenomenon.

Images