Fault-tolerant control method and device for flapping-wing aircraft based on high-gain observer
By adopting a fault-tolerant control method with high gain observer and non-singular fast terminal sliding mode surface design in the flapping aircraft, the problem of the flapping aircraft being susceptible to wind disturbances and actuator failures is solved, and the stable tracking of attitude and position is achieved Control improves the stability and reliability of the system.
Patent Information
- Application Number
- CN202211305282.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-24
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2042-10-24
AI Technical Summary
Due to its nonlinear, uncertainty, underdrive and strong coupling characteristics, the flapping wing aircraft is susceptible to wind disturbances and actuator failures, which makes it difficult to control attitude and position tracking, and the angular velocity is difficult to measure, affecting control performance.
Using a non-singular fast terminal sliding mode surface design based on high-gain observer, a mathematical model of attitude and position is obtained. The fault-tolerant controller is constructed through a high-gain observer and a non-singular fast terminal sliding mode surface to realize the attitude and attitude of the flapping aircraft. Position tracking control reduces system costs and improves reliability.
The attitude and position tracking control stability of the flapping wing aircraft is realized, and the system stability recovery is accelerated after failure, which reduces the cost of angular velocity measurement and improves the portability and reliability of the aircraft.
Smart Images

Figure CN115562327B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle control, and in particular to a fault-tolerant control method and device for a flapping-wing aircraft based on a high-gain observer. Background Art
[0002] The search formula is TACD_ALL: (flapping-wing aircraft AND adaptation AND control AND disturbance), and the relatively close prior art solutions are as follows.
[0003] The authorization announcement number is CN106707749B, and the name is a control method for a bionic flapping-wing flying robot, hereinafter referred to as comparative document 1. The method includes: performing kinematic and dynamic analysis on the bionic flapping-wing flying robot and establishing a Lagrangian model; based on the Lagrangian model, designing a model-based attitude controller with a disturbance observer; based on the designed model-based attitude controller with a disturbance observer, designing a neural network full-state attitude controller with a disturbance observer; based on the position control model of the bionic flapping-wing flying robot, designing a model-based position controller with a disturbance observer; according to the designed model-based attitude controller with a disturbance observer, the neural network full-state attitude controller and the model-based position controller, the trajectory tracking control of the attitude and position of the bionic flapping-wing flying robot is performed.
[0004] The authorization announcement number is CN113504729B, and the name is LMI-based H∞ robust control method and system for bionic flapping-wing aircraft, hereinafter referred to as comparative document 2. It includes: establishing a time-varying longitudinal dynamic model of a bionic flapping-wing aircraft; periodically averaging the aerodynamic force and torque of the bionic flapping-wing aircraft to obtain a nonlinear bionic flapping-wing aircraft longitudinal steady-state model; linearizing the nonlinear longitudinal dynamic steady-state model at the equilibrium point to obtain a linear nominal model; modeling the noise, error and actuator dynamic characteristics of the bionic flapping-wing aircraft according to the actual control process, and designing a weight function according to the amplitude and frequency characteristics of the noise, the amplitude attenuation gain and frequency characteristics of the error, and the frequency characteristics of the actuator dynamic characteristics; obtaining a generalized state space model of the controlled object based on the linear nominal model and the weight function, and converting the bionic flapping-wing aircraft closed-loop system into a standard form for solving the H∞ controller; obtaining a controller that satisfies robust stability and robust performance by solving linear matrix inequalities based on the LMI method.
[0005] Combining the above two patent documents and existing technical solutions, the inventors analyzed and found that the benefits of the existing technical solutions are:
[0006] In Comparative Document 1, only fewer sensors and actuators are needed to effectively track and control the posture and position of the bionic flapping-wing flying robot.
[0007] In Comparative Document 2, the bionic flapping-wing aircraft is transformed into a broad H∞ control framework, and a highly robust controller is obtained by solving multiple linear matrix inequalities, thereby realizing longitudinal control of the system in the presence of parameter uncertainty and noise disturbances; for aircraft systems with time-varying flight trajectories, the designed robust control rate is time-varying and has autonomous adjustment capabilities.
[0008] In recent years, driven by the development needs of national defense industry technology, the development of unmanned aerial vehicle technology has received great attention from major countries in the world, and bionic flapping-wing aircraft technology has therefore been greatly developed. Compared with fixed-wing aircraft and rotary-wing aircraft, flapping-wing aircraft have the advantages of small size, low energy consumption, strong maneuverability, and good bionic concealment, and can play a vital role in military, geological exploration, emergency rescue, etc.
[0009] Although flapping-wing aircraft have unique advantages in practical applications, they are not widely used in practice. The reason is that flapping-wing aircraft is a typical nonlinear, uncertain, under-driven, and strongly coupled system, which is susceptible to wind disturbances and actuator failures, all of which may lead to catastrophic consequences. In addition, the angular velocity of flapping-wing aircraft is difficult to measure. Attitude control that directly uses attitude measurements is widely used, but the approximate control-oriented model is still complex and has unknown parameters. All these characteristics increase the difficulty of accurate and robust attitude control. Therefore, for the safety and stability of flapping-wing aircraft control, it is of great significance to study the fault-tolerant control method of flapping-wing aircraft.
[0010] The above mentioned uncertainties, wind disturbances, actuator failures and no angular velocity control are detailed as follows:
[0011] Uncertainty: When a flapping-wing aircraft flies in different flight environments and flight modes, parameters such as the aerodynamic lift coefficient β1 and the damping coefficient β2 of the aircraft will inevitably change. The changes in these parameters will produce uncertain torques M on the aircraft. When performing attitude control of the aircraft, the uncertainty of these torques cannot be ignored.
[0012]
[0013] In formula (1), M1 and M2 represent the aerodynamic moment and damping moment of the body. f(β1) and f(β2) represent the function about β1 and the function about β2.
[0014] External wind disturbance: When a drone is in use, its flight motion is an extremely complex dynamic process. The existence of the wind field greatly affects the accuracy of its flight trajectory and flight attitude. Micro flapping-wing drones are more susceptible to wind disturbances due to their light weight and low speed.
[0015] Actuator failure: In actual control, actuators, sensors, and other communication or connection mechanisms may fail during operation. For example, the actuator motor of a flapping-wing drone fails, or the flapping wing is damaged during flight and can no longer deflect. These failures may lead to catastrophic consequences. Actuator failures can be divided into partial failure, complete failure, and loss of control. The voltage of a single motor is lost, which is equivalent to the normal operation of the motor but the flapping wing is damaged. Complete failure means that the motor voltage is zero or the flapping wing is stuck, and the lift and torque are zero. Loss of control means that the motor speed suddenly reaches the maximum and cannot be adjusted. At present, complete failure and loss of control can only be tolerated by redundant actuators. This article will not consider them for the time being. This article only considers failure failures and paranoid failures. The two types of failures are as follows:
[0016]
[0017] In formula (2), and represent attitude actuator failure and position actuator failure respectively, q r and q t represents the system attitude state matrix, and Indicates q r and q t The first derivative with respect to time t, u r and u t They represent attitude control input and position control input respectively. Formula (2) indicates that the flapping-wing aircraft system has a failure fault at 5-7s and a paranoid fault at 11-13s.
[0018] No angular velocity control: Flapping-wing aircraft can perform very complex movements, which makes it difficult to accurately measure the angular velocity, increasing the system cost. The measured angular velocity may contain a large amount of noise information, which will affect the control performance. In order to reduce costs and increase portability, some micro-aircraft are not equipped with angular velocity measurement equipment.
[0019] In the face of the above challenges in controller design, fruitful research results have been achieved in the past few decades:
[0020] To address the problem caused by actuator input saturation, an adaptive dynamic surface trajectory tracking control strategy with a saturation auxiliary system is proposed, which can effectively suppress input saturation. However, it does not study possible partial failures, bias faults, etc. of the actuator.
[0021] An active decentralized fault-tolerant control method using signal reconstruction is proposed. The fault signal is reconstructed through a numerical integrator and a differential tracker, and combined with an adaptive fuzzy decentralized control method to achieve fault-tolerant control of the robotic arm system. However, the influence of uncertainties such as interference and model errors is not considered.
[0022] In response to the situation where the actuator fails, an active fault-tolerant control scheme based on non-singular terminal sliding mode is proposed, using an adaptive extended Kalman filter as the fault detection and diagnosis module. Although this method has good fault tolerance, it requires measuring the angular velocity of the system, which increases the system cost. The measured angular velocity may contain a lot of noise information, which will affect the control performance.
[0023] Based on the existing technical solutions, the inventor hopes to find a new technical route and obtain a technical solution with independent intellectual property rights.
[0024] Existing technical problems and considerations:
[0025] How to solve the technical problems of attitude and position tracking control of flapping-wing aircraft. Summary of the invention
[0026] The technical problem to be solved by the present invention is to provide a fault-tolerant control method and device for a flapping-wing aircraft based on a high-gain observer, so as to solve the technical problem of attitude and position tracking control of the flapping-wing aircraft.
[0027] In order to solve the above technical problems, the technical solution adopted by the present invention is: a fault-tolerant control method for a flapping-wing aircraft based on a high-gain observer includes the steps of fault-tolerant control, obtaining an attitude mathematical model, obtaining a position mathematical model, obtaining a high-gain observer, and obtaining a non-singular fast terminal sliding surface, wherein the non-singular fast terminal sliding surface includes a non-singular fast terminal sliding surface in the attitude system and a non-singular fast terminal sliding surface in the position system; based on the attitude mathematical model, the high-gain observer and the non-singular fast terminal sliding surface in the attitude system, an attitude fault-tolerant controller based on a high-gain observer is obtained; based on the position mathematical model and the non-singular fast terminal sliding surface in the position system, a position fault-tolerant controller is obtained; and according to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer, attitude and position tracking control of the flapping-wing aircraft is performed.
[0028] A further technical solution is:
[0029]
[0030] In formula (16), u r1 is an attitude fault-tolerant controller based on a high-gain observer, 1<r2<2, v is a positive real number, s is a non-singular fast terminal sliding surface in the attitude system, z r1is the system angle error vector, z r1 T Represented as z r1 The transpose of r2 is the system angular velocity error vector, For z r2 The estimate of the gain matrix K r3 is a diagonal matrix, represents the estimate of the uncertainty of the flapping-wing aircraft system and the actuator failure, is the basis function of the neural network, d re is the attitude system disturbance, is the upper bound of the attitude system disturbance, and the adaptive parameter Express The estimate, is the elimination term in the attitude fault-tolerant controller based on the high-gain observer;
[0031]
[0032] In formula (23), u t1 is the position fault-tolerant controller, R IB is the coordinate transformation matrix, that is, the rotation matrix R BI (q r The inverse matrix of (t)), s t is the non-singular fast terminal sliding surface in the position system, z t1 is the system position error vector, z t1 T For z t1 The transpose of t2 is the system velocity error vector, the gain matrix K t3 is a diagonal matrix, M t (x t1 ) is the mass matrix, μ is a positive real number, r1>r2, Indicates z t1 The first derivative with respect to time t is is the approaching term in the non-singular fast terminal sliding surface derivative of the position system, is the elimination term in the fault-tolerant position controller, G is the weight matrix, α t is the virtual control input, To estimate the actuator failure of the flapping-wing aircraft position subsystem, is the upper bound of the position system disturbance, and the adaptive parameter for Estimates.
[0033] A further technical solution is that the posture mathematical model is:
[0034]
[0035] In formula (9), q r is the angular output of the attitude system, M r (q r ) is the inertia matrix, for q r The second derivative with respect to time t, is the centripetal force matrix, for q r The first derivative with respect to time t, u r is the attitude controller, d r is the external disturbance in the attitude system, It is an actuator failure in the attitude system.
[0036] A further technical solution is that the position mathematical model is:
[0037]
[0038] In formula (10), qt is the position output of the position system, is the velocity output in the position system, M t (q t ) is the mass matrix, for q t The second derivative with respect to time t, G t is the gravity matrix, R IB is the coordinate transformation matrix, q r is the attitude angle vector, u t is the position controller, d t is the position system disturbance, An actuator failure in the position system.
[0039] A fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer comprises a fault-tolerant control module, which is a program module and is used to obtain an attitude mathematical model, a position mathematical model, a high-gain observer, and a non-singular fast terminal sliding surface, wherein the non-singular fast terminal sliding surface comprises a non-singular fast terminal sliding surface in an attitude system and a non-singular fast terminal sliding surface in a position system; an attitude fault-tolerant controller based on a high-gain observer is obtained based on the attitude mathematical model, the high-gain observer, and the non-singular fast terminal sliding surface in the attitude system; a position fault-tolerant controller is obtained based on the position mathematical model and the non-singular fast terminal sliding surface in the position system; and attitude and position tracking control of a flapping-wing aircraft is performed according to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer.
[0040] A fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer comprises a memory, a processor and a computer program stored in the memory and executable on the processor. The processor implements the above corresponding steps when executing the computer program.
[0041] A fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer comprises a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above corresponding steps are implemented.
[0042] The beneficial effects of adopting the above technical solution are:
[0043] First, a fault-tolerant control method for a flapping-wing aircraft based on a high-gain observer includes the steps of fault-tolerant control, obtaining an attitude mathematical model, obtaining a position mathematical model, obtaining a high-gain observer, and obtaining a non-singular fast terminal sliding surface, wherein the non-singular fast terminal sliding surface includes a non-singular fast terminal sliding surface in the attitude system and a non-singular fast terminal sliding surface in the position system; obtaining an attitude fault-tolerant controller based on a high-gain observer based on the attitude mathematical model, the high-gain observer, and the non-singular fast terminal sliding surface in the attitude system; obtaining a position fault-tolerant controller based on the position mathematical model and the non-singular fast terminal sliding surface in the position system; and performing attitude and position tracking control on the flapping-wing aircraft according to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer. This technical solution, through the steps of fault-tolerant control, etc., realizes good stability in attitude and position tracking control of the flapping-wing aircraft, and can also accelerate the recovery of system stability after a fault. A high-gain observer is designed to estimate the angular velocity, which reduces the cost while improving the portability and reliability of the flapping-wing aircraft.
[0044] Second, a fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer includes a fault-tolerant control module, which is a program module for obtaining an attitude mathematical model, a position mathematical model, a high-gain observer, and a non-singular fast terminal sliding surface, wherein the non-singular fast terminal sliding surface includes a non-singular fast terminal sliding surface in the attitude system and a non-singular fast terminal sliding surface in the position system, an attitude fault-tolerant controller based on a high-gain observer is obtained based on the attitude mathematical model, the high-gain observer, and the non-singular fast terminal sliding surface in the attitude system, a position fault-tolerant controller is obtained based on the position mathematical model and the non-singular fast terminal sliding surface in the position system, and an attitude and position tracking control of the flapping-wing aircraft is performed according to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer. This technical solution, through the fault-tolerant control module, etc., realizes good stability in attitude and position tracking control of the flapping-wing aircraft, and can also accelerate the recovery of system stability after a fault, designs a high-gain observer to estimate the angular velocity, reduces the cost, and improves the portability and reliability of the flapping-wing aircraft.
[0045] Please refer to the detailed description of the specific implementation method. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 is a flow chart of Embodiment 1 of the present invention;
[0047] Figure 2 is a principle block diagram of embodiment 3 of the present invention;
[0048] Figure 3 is a principle block diagram of embodiment 4 of the present invention;
[0049] Figure 4 is a comparison diagram of the attitude trajectory error of a flapping-wing aircraft under the present invention and the previous control method;
[0050] Figure 5 It is a comparison diagram of the position trajectory error of flapping-wing aircraft under the control method of the present invention and the previous control method. DETAILED DESCRIPTION
[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.
[0053] Embodiment 1:
[0054] like Figure 1 As shown, the present invention discloses a fault-tolerant control method for a flapping-wing aircraft based on a high-gain observer, including the steps of fault-tolerant control, which are specifically divided into the following steps:
[0055] The first step is to obtain a mathematical model of attitude, a mathematical model of position, a high-gain observer, and a non-singular fast terminal sliding surface, wherein the non-singular fast terminal sliding surface includes a non-singular fast terminal sliding surface in the attitude system and a non-singular fast terminal sliding surface in the position system.
[0056] In the second step, an attitude fault-tolerant controller based on a high-gain observer is obtained based on the attitude mathematical model, the high-gain observer and the non-singular fast terminal sliding surface in the attitude system.
[0057] In the third step, a position fault-tolerant controller is obtained based on the position mathematical model and the non-singular fast terminal sliding surface in the position system.
[0058] In the fourth step, the attitude and position tracking control of the flapping-wing aircraft is performed according to the attitude fault-tolerant controller and position fault-tolerant controller based on the high-gain observer.
[0059] Embodiment 2:
[0060] The invention discloses a fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer, comprising a fault-tolerant control module based on a high-gain observer. The fault-tolerant control module designed based on the high-gain observer is a program module, which is used to obtain a posture mathematical model, a position mathematical model, a high-gain observer, and a non-singular fast terminal sliding surface. The non-singular fast terminal sliding surface comprises a non-singular fast terminal sliding surface in a posture system and a non-singular fast terminal sliding surface in a position system. Based on the posture mathematical model, the high-gain observer and the non-singular fast terminal sliding surface in the posture system, an attitude fault-tolerant controller based on the high-gain observer is obtained. Based on the position mathematical model and the non-singular fast terminal sliding surface in the position system, a position fault-tolerant controller is obtained. According to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer, attitude and position tracking control is performed on the flapping-wing aircraft.
[0061] Embodiment 3:
[0062] like Figure 2 As shown, the present invention discloses a fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of Example 1 are implemented.
[0063] Embodiment 4:
[0064] like Figure 3 As shown, the present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in Example 1 are implemented.
[0065] The technical problems to be solved are:
[0066] The problem to be solved by the present invention is the attitude and position tracking control problem of a flapping-wing aircraft with actuator failure and difficult-to-measure angular velocity.
[0067] Technical solution:
[0068] (1) The present invention studies the attitude and position trajectory tracking control under actuator failure, taking into account the actuator's paranoid failure and failure failure, so that the flapping-wing aircraft can still fly along the desired trajectory when the actuator fails.
[0069] (2) The present invention adopts a non-singular fast terminal sliding surface. Compared with the traditional sliding mode control, the non-singular fast terminal sliding surface can further improve the stability of the system and accelerate the recovery of the system stability after a fault.
[0070] (3) The present invention studies the non-angular velocity control and designs a high-gain observer to ensure that the angular velocity information is obtained when the angular velocity is difficult to measure.
[0071] (4) The present invention studies the attitude and position trajectory tracking control under wind disturbances, and adopts an adaptive law to estimate the upper bound of wind disturbances. Compared with the previous disturbance observer, the control method of the present invention can track the desired trajectory faster.
[0072] Technical solution description:
[0073] In order to solve the attitude and position tracking control problem of a fault-tolerant flapping-wing aircraft whose angular velocity is difficult to measure, a mathematical model of the attitude and position of a flapping-wing aircraft with faults, a high-gain observer, and a non-singular fast terminal sliding mode are designed. The design process is as follows:
[0074] The movement of a flapping-wing aircraft in the air is divided into translation and rotation around the center of mass of the aircraft. According to the principles of theoretical mechanics, the dynamic equation of the rotation of the aircraft around the center of mass is:
[0075]
[0076] In formula (3), the total moment of the aerodynamic moment and damping moment of the M body acting on the center of mass of the body, γ is the momentum moment acting on the center of mass of the body, which can be expressed as:
[0077] γ=I p ω (4)
[0078] In formula (4), I p is the moment of inertia matrix, ω=[pqr] T is the angular velocity of the inertial coordinate system after the three Euler angles of the body coordinate system are transformed, γ, M, ω are all vectors. p for:
[0079]
[0080] According to the calculation method of vector derivatives in the rotating coordinate system, the dynamic equation for rotation around the center of mass is:
[0081]
[0082] Substituting formula (4) into the formula, we can get the Lagrangian model as follows:
[0083]
[0084] In formula (7), T is the coordinate transformation matrix from the body coordinate system to the inertial coordinate system. r (q r )=I p T, and M r (q r ) has no practical meaning and is used to simplify the expression of the Lagrangian model; the system attitude state matrix q r (t) = [θ1 θ2 θ3] T , θ1, θ2, θ3 are the three Euler angles in the body coordinate system, namely roll angle, pitch angle, and yaw angle, is the system attitude state matrix q r (t) is the second-order derivative of time t, representing the angular velocity state matrix, Represents the system attitude state matrix q r (t) The second-order derivative with respect to time t, u r represents the attitude control input of the flapping-wing aircraft, and the coordinate transformation matrix T from the body coordinate system to the inertial coordinate system is expressed as:
[0085]
[0086] According to equations (2) and (7), the mathematical model of the attitude of a flapping-wing aircraft with a fault can be obtained as follows:
[0087]
[0088] In formula (9), d r It is the external disturbance in the rotational motion state.
[0089] The translational motion equation of a flapping-wing aircraft, i.e., the position mathematical model, is as follows:
[0090]
[0091] In formula (10), is the mass matrix, is the gravity matrix and coordinate transformation matrix,
[0092]
[0093] And R IB (qr(t))=(R BI (qr(t)))- 1 ;qt=[xyz]T represents the position state matrix, is the first-order derivative of qt with respect to time t, representing the velocity vector, for q t The second-order derivative with respect to time t, m is the total mass of the flapping-wing aircraft, u t represents the position controller, d t is the system disturbance, Actuator failure.
[0094] Among them, the disturbance d of the system in equations (9) and (10) is r =[d r1 d r2 d r3 ] T , d t =[d t1 d t2 d t3 ] T is bounded, satisfying: in, and is an unknown constant, and is a skew-symmetric matrix. Let x r1 =q r (t) = [θ1 θ2 θ3] T , Used to represent the state vector of the system, representing the angle and angular velocity of the flapping-wing aircraft, x r1 is the output of the attitude system, and the state equation of the attitude system of the flapping-wing aircraft is obtained:
[0095]
[0096] In formula (11), M r -1 (x r1 ) is M r (x r1 ), and They are x r1 and x r2 The first derivative with respect to time t.
[0097] Since flapping-wing aircraft can perform very complex motions, this gives the angular velocity state vector x r2 The accurate measurement of the angular velocity state vector x r2 It may contain a lot of noise information, which will affect the control performance. The present invention designs a high-gain observer to estimate the angular velocity state vector x which is difficult to measure. r2 Information. The design is as follows:
[0098] According to the definition of a high-gain observer, the unmeasurable state x r2 The estimated design is:
[0099]
[0100] Tracking error term z r2 The estimate of can be expressed as:
[0101]
[0102] The dynamic equation of π2 is expressed as:
[0103]
[0104] where δ is an arbitrarily small positive constant, The choice is to let becomes a Hurwitz polynomial, then the following properties hold η (k) represents the k-th derivative of η, And for any Both ||ξ k ||≤δh k .
[0105] In order to improve the stability of the system and accelerate the recovery of system stability after a fault, the non-singular fast terminal sliding mode surface of the attitude system is constructed as follows according to the above high-gain observer (12):
[0106]
[0107] Where s is the constructed non-singular fast terminal sliding surface, z r1 =x r1 -x rd ,z r2 =x r2 -α r are the angle error vector and the angular velocity error vector, respectively. Represents the state angular velocity error vector z r2 The estimate of x rd =[x r1d x r2d x r3d ] T is the expected angle value of the flapping-wing UAV, α r =[α r1 α r2 α r3 ] T is the virtual control input, μ>0, v>0, 1<r2<2, r1>r2, e,f,g,h are odd numbers.
[0108] From the above state equation (11), it can be seen that the output state vector of the attitude system of the present invention is x r1 and x r2 , the angular state vector x r1 Easy to measure, so the controller is designed to be r1 Known x r2 The attitude fault-tolerant controller designed under unknown conditions can be designed based on the state equation (11), the high-gain observer (12) and the non-singular fast terminal sliding surface (15) as follows:
[0109]
[0110] In formula (16), u r1 For the attitude fault-tolerant controller designed in this paper, the gain matrix K r3 is a diagonal matrix, and the elements on the diagonal are greater than 1, Represents the estimation of the uncertainty of the flapping-wing aircraft system and the actuator failure, and the adaptive parameter Expressed as the ideal neural network weight W * The estimated value of W * =[W * 1 W * 2 W * 3],W * i =[W * i1 W * i2 W * i3 W * i4 W * i5 ],i=1,2,3, is the basis function of the neural network, in represents the input of the neural network, c ij =c 1j c 2j c 3j c 4j c 5j represents the kernel parameter of the Gaussian function; b represents the width of the Gaussian function; exp(·) represents the exponential function with e as the base, and the adaptive parameter Express The estimate,
[0111] The adaptive parameters are calculated by the following adaptive law: and
[0112]
[0113] In formula (17), the design parameter Γ dre ,Γ W ,σ dre , σ W >0.
[0114] The controller (16) designed as above can control the attitude angle q of the flapping-wing aircraft during flight. r , substituting the attitude fault-tolerant controller (16) designed based on the high-gain observer into the attitude state equation (11) of the flapping-wing aircraft, we can obtain:
[0115]
[0116] In order to verify the stability of the designed controller (17), the Lyapunov function V1 is constructed:
[0117]
[0118] In formula (19),
[0119] By taking the derivative of formula (19), we can get:
[0120]
[0121] In formula (20), Gain matrix K r1 is a diagonal matrix, and K r1 The diagonal elements are greater than I is the identity matrix, is a normal number, is the approximation error vector, For the radial basis function of the neural network based on Gaussian function, represents the observed value of the radial basis function input, ψ represents the bounded constant error vector, and we can get where ε t is a constant.
[0122] Considering the estimation error of the observer The influence of , construct Lyapunov function according to η (k) represents the k-th derivative of η, And for any Both ||ξ k ||δh k , it can be seen that V obsis stable and can be written as Among them, ξ1, h1, is a positive constant.
[0123] In summary, the system is stable as long as the Lyapunov function V1 is stable. λ min (K r3 -I)>0, the closed-loop system is eventually uniformly bounded and stable.
[0124] Similarly, the state equation of the flapping-wing aircraft position system is established according to equation (10), and x t1 =q t1 =[xyz] T , x t1 and x t2 is the output of the position system, and the state equation is:
[0125]
[0126] In formula (21), M t -1 (x t1 ) is M t (x t1 ), and They are x t1 and x t2 The first derivative with respect to time t.
[0127] Since the output quantity x in the position system t1 , x t2 Since they are easy to measure, the state observer is no longer designed. In order to improve the stability of the system and accelerate the recovery of system stability after a fault, a non-singular fast terminal sliding surface of the position system is constructed:
[0128]
[0129] Among them, z t1 =x t1 -x td ,z t2 =x t2 -α t are the position error vector and the velocity error vector respectively, Indicates z t1 The first derivative with respect to time t, where x td =[x t1d x t2d x t3d ] T is the expected angle value of the flapping-wing UAV, α t =[α t1α t2 α t3 ] T is the virtual control input, μ>0, v>0, 1<r2<2, r1>r2, e,f,g,h are odd numbers.
[0130] According to the above state equation, the position fault-tolerant controller can be designed as:
[0131]
[0132] In formula (23), u t1 For the position fault-tolerant controller designed in this paper, a⊙b=[a1b1,a2b2,...,a n b n ] T , the gain matrix K t3 is a diagonal matrix, and the elements on the diagonal are greater than represents the estimation of actuator failure of the flapping-wing aircraft position subsystem, and the adaptive parameter Expressed as the ideal neural network weight W t * The estimated value of W t * =[W * t1 W * t2 W * t3 ],W * ti =[W * ti1 W * ti2 W * ti3 W * ti4 W * ti5 ], i=1,2,3, S(X)=[S1, S2, S3, S4, S5] T is the basis function of the neural network, Where X = [z r1 T ,z r2 T ,s t T ,u t T ] T represents the input of the neural network, c ij =c 1j c 2j c 3j c 4j c5j represents the kernel parameter of the Gaussian function; b represents the width of the Gaussian function; exp(·) represents the exponential function with e as the base, and the adaptive parameter for The estimate,
[0133] The adaptive parameters are calculated by the following adaptive law: and
[0134]
[0135] In formula (24), the design parameter Γ dte ,Γ Wt ,σ dte , σ Wt >0.
[0136] The controller (23) designed as above can control the position state q of the flapping-wing aircraft during flight. t , Substituting the position fault-tolerant controller (23) into the position state equation (21) of the flapping-wing aircraft, we can obtain:
[0137]
[0138] In order to verify the stability of the designed controller, the Lyapunov function V2 is constructed:
[0139]
[0140] In formula (26),
[0141] By taking the derivative of formula (26), we can get:
[0142]
[0143] In formula (27), Gain matrix K t1 is a diagonal matrix, and K t1 The elements on the diagonal are greater than 0, I is the identity matrix, is a normal number, ε(X)=[ε1,ε2,ε3] T is the approximation error vector,
[0144] In summary, when the conditions are met When , the closed-loop system is eventually uniformly bounded and stable.
[0145] In order to prove the effectiveness of the attitude fault-tolerant controller and position fault-tolerant controller designed based on the high-gain observer, the present invention uses MATLAB to establish a flapping-wing aircraft system, and the parameters are selected as shown in Table 1:
[0146] Table 1: Flapping-wing micro-aircraft parameters
[0147]
[0148] Assume that the desired trajectory of the posture is selected as The expected trajectory of the position is selected as The disturbances for attitude control and position control are Actuator failure in attitude control is Actuator fault in position control is It means that a failure occurs at 5-7s, and a paranoid failure occurs at 11-13s. The initial value of the state variable is set to: x r1 =[0.8,-0.5,0.5] T mm, x r2 =[0.1,0.1,0.1] T mm, x t1 =[0.5,0.5,0.5] T mm, x t2 =[0.1,0.1,0.1] T mm. The design parameters of the attitude controller, position controller and observer parameters are set as follows: K r1 =[4,0,0;0,4,0;0,0,4],K r3 =[20,0,0;0,20,0;0,0,20],K t1 =[30,0,0;0,30,0;0,0,40],K t3 =[40,0,0;0,40,0;0,0,40]. σ W =0.2,Γ W =2. dr =10,Γ dr =0.08,σ Wt =0.2,Γ Wt =2,σ dt =0.02,Γ dt =1,δ=0.004, τ=0.1, μ=300, v=6, e=5, f=3, g=9, h=7.
[0149] like Figure 4 and Figure 5As shown, it can be seen that compared with the previous control methods that did not consider actuator failures, the control strategy designed in this paper can not only effectively deal with the adverse effects of flapping-wing aircraft caused by actuator failures during flight, but also can quickly track the desired attitude and position trajectory.
[0150] After this application was run internally for a period of time, the on-site technicians reported that the application was beneficial in the following ways:
[0151] Compared with the current research results, the main contributions of the present invention are mainly reflected in three aspects. First, when an actuator failure occurs in a flapping-wing aircraft, it can not only effectively eliminate the impact of the failure, but also track the expected trajectory. Secondly, the controller designed by the present invention adds a non-singular fast terminal sliding surface, which can not only further improve the stability of the system, but also accelerate the recovery of the stability of the system after the failure. Finally, the present invention designs a high-gain observer to estimate the angular velocity, which reduces the cost while improving the portability and reliability of the flapping-wing aircraft.
[0152] At present, the technical solution of the present invention has been put into pilot production, that is, a small-scale test of the product before large-scale mass production; after the pilot production was completed, a user usage survey was carried out in a small range, and the survey results showed that user satisfaction was high; now preparations have begun for the formal production of the product for industrialization (including intellectual property risk warning survey).
Claims
1. A fault-tolerant control method for a flapping-wing aircraft based on a high-gain observer, characterized in that: The method comprises the steps of fault-tolerant control, wherein the steps of the fault-tolerant control comprise: obtaining an attitude mathematical model, obtaining a position mathematical model, obtaining a high-gain observer, and obtaining a non-singular fast terminal sliding surface, wherein the high-gain observer is used to estimate the angular velocity, and the non-singular fast terminal sliding surface comprises a non-singular fast terminal sliding surface in the attitude system and a non-singular fast terminal sliding surface in the position system; obtaining an attitude fault-tolerant controller based on the high-gain observer based on the attitude mathematical model, the high-gain observer, and the non-singular fast terminal sliding surface in the attitude system; obtaining a position fault-tolerant controller based on the position mathematical model and the non-singular fast terminal sliding surface in the position system; and performing attitude and position tracking control on a flapping-wing aircraft according to the attitude fault-tolerant controller and the position fault-tolerant controller based on the high-gain observer; In formula (15), s is the non-singular fast terminal sliding surface in the constructed attitude system, z r1 =x r1 -x rd ,z r2 =x r2 -α r are the angle error vector and the angular velocity error vector, respectively. Represents the state angular velocity error vector z r2 The estimate of x rd =[x r1d x r2d x r3d ] T is the expected angle value of the flapping-wing UAV, α r =[α r1 α r2 α r3 ] T is the virtual control input, μ>0, v>0, 1<r2<2, r1>r2, e, f, g, h are odd numbers; In formula (16), u r1 is an attitude fault-tolerant controller based on a high-gain observer, 1<r2<2, v is a positive real number, s is a non-singular fast terminal sliding surface in the attitude system, z r1 is the system angle error vector, z r1 T Represented as z r1 The transpose of r2 is the system angular velocity error vector, For z r2 The estimate of the gain matrix K r3 is a diagonal matrix, represents the estimate of the uncertainty of the flapping-wing aircraft system and the actuator failure, is the basis function of the neural network, d re is the attitude system disturbance, is the upper bound of the attitude system disturbance, and the adaptive parameter Express The estimate, is the elimination term in the attitude fault-tolerant controller based on the high-gain observer; In formula (22), s t is the non-singular fast terminal sliding surface in the position system, z t1 =x t1 -x td ,z t2 =x t2 -α t are the position error vector and the velocity error vector respectively, Indicates z t1 The first derivative with respect to time t, x td =[x t1d x t2d x t3d ] T is the desired position of the flapping-wing UAV, α t =[α t1 α t2 α t3 ] T is the virtual control input, μ>0, v>0, 1<r2<2, r1>r2, e, f, g, h are odd numbers; In formula (23), u t1 is the position fault-tolerant controller, R IB is the coordinate transformation matrix, that is, the rotation matrix R BI (q r (t)), the inverse matrix of the system attitude state matrix q r (t) = [θ1θ2θ3] T , θ1, θ2, θ3 are the three Euler angles in the body coordinate system, namely roll angle, pitch angle, yaw angle, s t is the non-singular fast terminal sliding surface in the position system, z t1 is the system position error vector, z t1 T For z t1 The transpose of t2 is the system velocity error vector, the gain matrix K t3 is a diagonal matrix, M t (x t1 ) is the mass matrix, μ is a positive real number, r1>r2, Indicates z t1 The first derivative with respect to time t is is the approaching term in the non-singular fast terminal sliding surface derivative of the position system, is the elimination term in the fault-tolerant position controller, G is the weight matrix, α t is the virtual control input, To estimate the actuator failure of the flapping-wing aircraft position subsystem, is the upper bound of the position system disturbance, and the adaptive parameter for Estimates.
2. The fault-tolerant control method for flapping-wing aircraft based on a high-gain observer according to claim 1, characterized in that: The mathematical model of posture is: In formula (9), q r is the angular output of the attitude system, M r (q r ) is the inertia matrix, Q r The second derivative with respect to time t, is the centripetal force matrix, Q r The first derivative with respect to time t, u r is the attitude controller, d r is the external disturbance in the attitude system, It is an actuator failure in the attitude system.
3. The fault-tolerant control method for flapping-wing aircraft based on a high-gain observer according to claim 1, characterized in that: The mathematical model of position is: In formula (10), q t is the position output in the position system, is the velocity output in the position system, M t (q t ) is the mass matrix, Q t The second derivative with respect to time t, G t is the gravity matrix, R IB is the coordinate transformation matrix, q r is the attitude angle vector, u t is the position controller, d t is the position system disturbance, An actuator failure in the position system.
4. A flapping-wing aircraft fault-tolerant control device based on a high-gain observer, used in the flapping-wing aircraft fault-tolerant control method based on a high-gain observer according to claim 1, characterized in that: The invention comprises a fault-tolerant control module, which is a program module and is used for obtaining a posture mathematical model, a position mathematical model, a high-gain observer, and a non-singular fast terminal sliding surface. The non-singular fast terminal sliding surface comprises a non-singular fast terminal sliding surface in a posture system and a non-singular fast terminal sliding surface in a position system. Based on the posture mathematical model, the high-gain observer and the non-singular fast terminal sliding surface in the posture system, a posture fault-tolerant controller based on the high-gain observer is obtained. Based on the position mathematical model and the non-singular fast terminal sliding surface in the position system, a position fault-tolerant controller is obtained. According to the posture fault-tolerant controller based on the high-gain observer and the position fault-tolerant controller, the posture and position tracking control of the flapping-wing aircraft is performed.
5. A fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the corresponding steps of any one of claims 1 to 3 are implemented.
6. A fault-tolerant control device for a flapping-wing aircraft based on a high-gain observer, comprising a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, characterized in that: When the computer program is executed by a processor, the corresponding steps of any one of claims 1 to 3 are implemented.
Citation Information
Patent Citations
Robust Control Method and System for Bionic Flapping-Wing Aircraft Based on LMI (Limited Mind Machine)
CN113504729B
Control method for bionic flapping-wing aerial robot
CN106707749A
Self-adaptation trajectory tracking control method for autonomous underwater vehicle horizontal plane
CN108427414A
