Endogenous safety self-adaptive fault-tolerant control method of overdrive vector rotor aircraft

Abstract

The invention discloses an endogenous safety self-adaptive fault-tolerant control method for an overdrive vector rotor aircraft, solves the problem of how to improve the reliability of the vector rotor aircraft, and belongs to the field of rotor aircraft control. The method comprises the steps that in order to ensure that the vector rotorcraft can achieve a good fault-tolerant effect under different fault mode conditions, the fault state is described as the actuator aerodynamic force loss; the deviation of an expected control distribution matrix and an actual control distribution matrix at the current moment in a dynamic system and reference model output error of the control distribution system is progressively converged to a self-adaptive law of a zero design error, so that bounded uncertainty does not need to be estimated; the method only depends on the IMU feedback information of the aircraft, and no additional sensor needs to be added for fault detection. The vector rotor function can be realized under the conditions of complete failure of a single rotor, complete failure of two interphase and opposite rotors and complete failure of three interphase rotors.

Description

Technical Field

[0001] The present invention relates to an endogenic safety adaptive fault-tolerant control method for an over-driven vector rotorcraft, belonging to the field of rotorcraft control. Background Art

[0002] Due to characteristics such as simple structure, ability to take off and land vertically, and stable hovering, rotorcraft have been widely used in tasks such as aerial reconnaissance, mapping, and high-altitude inspection. However, traditional quadrotor UAVs with coplanar thrusters have under-actuated characteristics, resulting in poor performance in complex or force interaction scenarios. The biaxial vector rotorcraft introduces vector control technology on the basis of traditional rotorcraft, significantly improving the independent control ability of the aircraft's position and attitude and expanding the application scenarios of rotorcraft. However, the additional actuators pose a challenge to the reliability of the aircraft. Summary of the Invention

[0003] Aiming at the problem of how to improve the reliability of vector rotorcraft, the present invention provides an endogenic safety adaptive fault-tolerant control method for an over-driven vector rotorcraft.

[0004] An endogenic safety adaptive fault-tolerant control method for an over-driven vector rotorcraft of the present invention includes:

[0005] S1. Taking as the control allocation matrix deviation, and gradually converging to 0, designing the adaptive law of the error; e c (k) represents the output error between the dynamic system and the reference model at the k-th moment, Am represents the state matrix of the dynamic system, L is selected to satisfy the poles are located inside the unit circle, and θ(k) is the desired control allocation matrix at the current moment;

[0006] S2. According to the actual control allocation matrix at the current moment and the desired control quantity w c (k) at the current moment, obtaining the aerodynamic force command signal matrix at the current moment. After applying u ae / A (k) to the vector rotorcraft, obtaining the actual control quantity w'(k) = B c / A Λu ae / A (k);

[0007] where, B c / A represents the control efficiency matrix, respectively represent the three-axis components f xi , f yi , f ziThe loss ratio, where \(i = 1, 2, 3, 4, 5, 6\);

[0008] S3. According to the actual control allocation matrix at the current moment and the actual control quantity \(w'(k)\) and the desired control quantity \(w\) at the current moment c (k), the actual control allocation matrix at the next moment is obtained using the adaptive law of the error

[0009] S4. According to the actual control allocation matrix at the next moment and the desired control quantity \(w\) c (k + 1), the aerodynamic force command signal matrix at the next moment is obtained Fault-tolerant control is completed.

[0010] Preferably, the adaptive law of the error is:

[0011]

[0012] where \(\epsilon\) w (k) represents the aerodynamic force deviation, \(\mu(k)\) is the scaling coefficient, and \(\Gamma\) d is the control allocation adaptive rate matrix.

[0013] Preferably, the control quantity includes aerodynamic force and aerodynamic torque.

[0014] When \(u\) ae / A (k) acts on the vector rotorcraft, the angular acceleration of the vector rotorcraft is collected and the acceleration in the inertial frame According to the angular acceleration and the acceleration in the inertial frame the actual control quantity is calculated

[0015] where \(f'\) is the actual aerodynamic force and \(t'\) is the actual aerodynamic torque;

[0016]

[0017] \(m\) is the mass of the vector rotorcraft, \(J\) is the moment of inertia in the body coordinate system, \(r\) COM is the position of the center of mass, \(\omega\) is the angular velocity of rotation, \(R\) EB represents the rotation matrix of the inertial frame relative to the body coordinate system, \(R\) BE represents the rotation matrix of the body coordinate system relative to the inertial frame, \(f\) d represents the total disturbance force, \(t\) d represents the total disturbance torque, and \(g\) represents the gravitational acceleration constant.

[0018] Advantages of the present invention: The present invention is designed for the control allocator of a vector rotorcraft. When one or more vector propulsion units of the vector rotorcraft fail, the actual control allocation matrix is updated online to maintain the six-degree-of-freedom flight ability of the aircraft, thereby improving the reliability of the vector rotorcraft. The present invention can realize the vector rotor function in the case of complete failure of a single rotor, complete failure of two adjacent or opposite rotors, and complete failure of three adjacent rotors. Description of the Drawings

[0019] Figure 1 Schematic diagram of a dual-axis tilt vector rotorcraft;

[0020] Figure 2 Schematic diagram of the structure of the vector propulsion unit;

[0021] Figure 3 Overall control model of the vector rotorcraft;

[0022] Figure 4 Flowchart of model reference adaptive control allocation. Detailed Embodiment

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0024] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0025] Next, the present invention will be further described in conjunction with the accompanying drawings and specific embodiments, but it is not a limitation of the present invention.

[0026] The over-driven vector rotorcraft's endogenous safety adaptive fault-tolerant control method in this embodiment includes:

[0027] Step 1: Design the adaptive law of the error:

[0028] First, perform dynamic modeling of the vector propulsion unit. A dual-axis tilt vector six-rotorcraft is as shown in Figure 1 , and the schematic diagram of the structure of the vector propulsion unit is as shown in Figure 2 .

[0029] Establish the body coordinate system C B on the aircraft body, and the vector propulsion unit coordinate system C AiIt is fixed relative to the body coordinate system, and the x-axis lies in the Oxy plane of the body coordinate system. Suppose that at the initial position, the angle of rotation of the i-th vector propulsion unit and the body coordinate system about the z-axis direction is δ i , then their rotation relationship satisfies:

[0030]

[0031] where c(δ i ) represents cos(δ i ), and s(δ i ) represents sin(δ i );

[0032] Suppose the distance between the origin of the vector propulsion unit coordinate system and the body coordinate system is l AB , then their position relationship satisfies:

[0033]

[0034] As Figure 2 shown, i = 1, 2, 3, 4, 5, 6. Suppose the rotation angle of the servo in the i-th vector propulsion unit is α i ; the rotation angle of the outer servo is λ i ; the motor speed is The aerodynamic constant is c f . Then the generated vector thrust f i The expression in the vector propulsion unit coordinate system is:

[0035]

[0036] If the desired vector thrust f i is known, then the actuator command signal can be solved according to Equation (3):

[0037]

[0038] where f xi , f yi , f zi respectively represent the three-axis components of f i along the vector propulsion unit coordinate system;

[0039] Considering the anti-torque generated by the propeller, suppose the anti-torque constant of the propeller is c r , and the vector torque t i generated by the i-th vector propulsion unit in the body coordinate system is expressed as:

[0040]

[0041] Among them, cw(i) represents the rotation direction of the rotor. cw(i) = 1 indicates clockwise rotation, and cw(i) = -1 indicates counterclockwise rotation. represents the position vector of the coordinate system of the i-th vector propulsion unit in the body coordinate system. represents the skew-symmetric matrix of the position vector;

[0042] For the total aerodynamic force f ae / B and the aerodynamic torque t ae / B expressed in the body coordinate system as:

[0043]

[0044] Among them, B c / A represents the control efficiency matrix, and u ae / A represents the aerodynamic force command signal matrix;

[0045] The control efficiency matrix is defined as follows:

[0046]

[0047] The aerodynamic force command signal matrix is defined as follows:

[0048]

[0049] The control method of the vector rotorcraft often adopts hierarchical control as Figure 3 shown, Figure 3 where x t represents the desired state of the system, including position, attitude, velocity, and angular velocity, and x b represents the actual state of the system, represents the desired aerodynamic force f c and the aerodynamic moment t c , where the high-level controller realizes the trajectory tracking ability of the vector rotorcraft while estimating and compensating for disturbances, and calculates the desired aerodynamic force and aerodynamic moment from the desired trajectory. The control allocator then distributes the desired aerodynamic force and moment generated by the high-level controller to the command signals of each actuator. Since the failure of the vector propulsion unit does not affect the high-level controller to generate the desired aerodynamic force and aerodynamic moment, the design method of the high-level controller is not particularly described in this embodiment. In actual work, any controller design method can be used to obtain the desired aerodynamic force f c and the aerodynamic moment t c . In this embodiment, the design mainly focuses on the control allocator.

[0050] To ensure that the vector rotorcraft can achieve a better fault tolerance effect under different fault modes, the fault state is described as the loss of aerodynamic force of the actuator in this embodiment. Let respectively represent the loss ratios after the fault f xi , f yi , f zi . Then the aerodynamic force f i ' generated by the vector propulsion unit after the fault can be expressed as:

[0051]

[0052] Then the actual aerodynamic force f' and aerodynamic torque t′ generated after the fault are as follows:

[0053]

[0054] The pseudo-inverse method is used to solve the desired control allocation matrix. It is assumed that the number of completely failed vector rotors does not exceed 3, that is, B c / A is row full rank and the full drive characteristic of the system is guaranteed. It is assumed that the desired control allocation matrix θ can realize w'→w c The energy-optimal solution of the desired control allocation matrix θ is obtained from Equation (10):

[0055]

[0056] To realize the estimation of the desired control allocation matrix θ This patent adopts a model reference adaptive method based on the feedback of actual aerodynamic force to design the adaptive law.

[0057] At present, most adaptive methods are designed based on the Lyapunov law of continuous systems. However, the actual system controller is often a digital controller, and the discretization step will lead to the loss of stability margin and even make the control system unstable. Therefore, this patent designs the discrete adaptive control law based on sampled data and proves its convergence.

[0058] In this embodiment, it is assumed that the time constant of the actuator dynamic process is very small and can be ignored, that is, the actuator immediately responds to the command signal u generated by the control allocation ae / A . Then the actual aerodynamic force and aerodynamic moment w' should satisfy the following equation:

[0059]

[0060] To reduce the influence of model error on the control effect, in traditional adaptive methods, it is usually necessary to estimate the uncertainty in the system online, which usually slows down the ability of the control system to track fast-changing signals. In this embodiment, the virtual allocation state y(k) of the dynamic system and the control allocation state y m (k) of the reference model update the adaptive actual control allocation matrix through their error, so as to eliminate the need to estimate the bounded uncertainty.

[0061] The actual aerodynamic force and aerodynamic torque w'(k) and the high-level controller expected output w c (k) deviation is used as feedback, θ(k) is the expected control allocation matrix at the current moment, and the control allocation matrix deviation is defined as Assume A m The poles are located inside the unit circle, and the desired control matrix θ satisfies B c / A Λθ=I, and define the system as follows:

[0062]

[0063] Among them, A m represents the state matrix of the dynamic system, and y(k) represents the output of the dynamic system at time k;

[0064] The output deviation between the dynamic system and the reference model is used as feedback to form a closed loop model. The reference model is defined as follows while ensuring the convergence speed of the dynamic system without introducing too much oscillation:

[0065] y m (k+1)=A m y m (k)-L(y(k)-y m (k)) (14)

[0066] The selection of L satisfies The pole is located inside the unit circle, and the control distribution error e c (k)Satisfy e c (k) = y(k) - y m (k), we get the error system:

[0067]

[0068] In order to achieve the control allocation error asymptotically converging to 0 and thus to estimate the desired control allocation matrix θ, an adaptive law is designed for the error system.

[0069] In a preferred embodiment, the adaptive law of this implementation is designed as follows:

[0070]

[0071] Among them, ∈ w (k) represents the aerodynamic force deviation, μ(k) is the scaling factor, Γ d Assigning adaptive rate matrix to control

[0072] Defining a positive definite sequence a w Satisfy the inequality λ max Representing the maximum eigenvalue of the matrix gives:

[0073]

[0074] That is, the positive definite sequence V d (k) decreases, and further it can be deduced that:

[0075]

[0076] Sum both sides of equation (17) to obtain:

[0077]

[0078] Since |a w | 2 and μ(k) are both greater than zero, it can be deduced that:

[0079]

[0080] That is, the estimation error asymptotically converges to 0, and a good estimation of the desired control allocation matrix can be achieved:

[0081]

[0082] Step 2: According to the actual control allocation matrix at the current moment and the desired control quantity w at the current moment c (k), obtain the aerodynamic force command signal matrix at the current moment After applying u ae / A (k) to the vector rotor aircraft, obtain the actual control quantity w'(k);

[0083] When the aerodynamic force command signal matrix u at the current moment is obtained for the first time ae / A (0), is the set initial value, and at subsequent moments is updated according to the adaptive law of the error;

[0084] Different from active fault-tolerant control which requires adding additional sensors to detect actuator faults, the method of this embodiment only depends on the feedback information of the aircraft IMU and has a certain universality. As Figure 4 shown. After applying u ae / A (k) to the vector rotor aircraft, collect the angular acceleration and the acceleration in the inertial frame According to the angular acceleration and the acceleration in the inertial frame calculate the actual control quantity Neglecting the high-order dynamic effects brought by the rotor tilting, regarding the vector rotor aircraft as a rigid body, the actual aerodynamic force f' and the actual aerodynamic torque t′ are respectively:

[0085]

[0086] where m is the mass of the vector rotorcraft, J is the moment of inertia in the body coordinate system, r COM is the position of the center of mass, ω is the angular velocity of rotation, R EB represents the rotation matrix of the inertial system relative to the body coordinate system, R BE represents the rotation matrix of the body coordinate system relative to the inertial system, f d represents the total disturbing force, t d represents the total disturbing torque, and g represents the gravitational acceleration constant.

[0087] Step 3: According to the actual control allocation matrix at the current moment and the actual control quantity w'(k) and the desired control quantity w at the current moment c (k), use the adaptive law of the error in Equation (16) to obtain the actual control allocation matrix at the next moment

[0088] Step 4: According to the actual control allocation matrix at the next moment and the desired control quantity w at the next moment c (k + 1), obtain the aerodynamic command signal matrix at the next moment Complete one fault-tolerant control.

[0089] This embodiment only depends on the estimation of the actual aerodynamic force of the rotor by the state information of the aircraft itself as the feedback quantity, online estimates the control allocation matrix, and realizes the fault-tolerant function of the vector rotorcraft. This embodiment is based on the design of a fault-tolerant control allocator for a vector rotorcraft based on model reference adaptive theory, and can be combined with any high-level controller in practical applications to realize the fault-tolerant function of the vector rotorcraft.

[0090] Although the present invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed, as long as they do not depart from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the different dependent claims and the features described herein can be combined in a manner different from that described in the original claims. It should also be understood that the features described in connection with a single embodiment can be used in other described embodiments.

Claims

1. An intrinsically safe adaptive fault-tolerant control method for an overdriven vector rotorcraft, characterized in that: include: S1, As the control allocation matrix deviation, middle Asymptotically converges to 0, the adaptive law of design error; e c (k) represents the output error between the dynamic system and the reference model at time k, A m Represents the state matrix of the dynamic system, and L is selected to satisfy The poles are located inside the unit circle, and θ(k) is the desired control allocation matrix at the current moment; S2, according to the actual control allocation matrix at the current moment and the expected control quantity w at the current moment c (k) Get the current aerodynamic force command signal matrix will u ae / A (k) acts on the vector rotorcraft, and the actual control amount w'(k) = B is obtained c / A Λu ae / A (k); Among them, B c / A represents the control efficiency matrix, They represent the three-axis components f of the vector thrust in the i-th vector propulsion unit coordinate system after the i-th vector propulsion unit fails. xi 、f yi 、f zi The loss ratio, i = 1, 2, 3, 4, 5, 6; S3, according to the actual control allocation matrix at the current moment and the actual control amount w'(k) at the current moment and the expected control amount w at the current moment c (k), using the adaptive law of the error to obtain the actual control allocation matrix at the next moment S4. According to the actual control allocation matrix at the next moment and the expected control amount w at the next moment c (k+1), get the aerodynamic force command signal matrix for the next moment Complete fault-tolerant control.

2. The intrinsically safe adaptive fault-tolerant control method for an overdriven vector rotorcraft according to claim 1, characterized in that: The adaptive law of the error is: Among them, ∈ w (k) represents the aerodynamic force deviation, μ(k) is the scaling factor, Γ d Assign an adaptive rate matrix to the control.

3. The intrinsically safe adaptive fault-tolerant control method for an overdriven vector rotorcraft according to claim 1, characterized in that: The pseudo-inverse method is used to solve the desired control allocation matrix θ(k) at the current moment: θ(k) T =(B c / A L) T (B c / A L(B c / A L) T ) -1 。 4. The method for intrinsically safe adaptive fault-tolerant control of an overdriven vector rotorcraft according to claim 1, characterized in that: The controlled quantities include aerodynamic force and aerodynamic torque. u ae / A (k) After acting on the vector rotorcraft, the angular acceleration of the vector rotorcraft is collected and the acceleration in the inertial frame According to the angular acceleration and the acceleration in the inertial frame Calculate the actual control amount Among them, f' is the actual aerodynamic force, t' is the actual aerodynamic torque; m is the mass of the vector rotorcraft, J is the moment of inertia in the body coordinate system, r COM is the center of mass position, ω is the angular velocity, R EB Represents the rotation matrix of the inertial system relative to the body coordinate system, R BE represents the rotation matrix of the body coordinate system relative to the inertial system, f d represents the total disturbance force, t d represents the total disturbance torque, and g represents the gravitational acceleration constant.

5. The intrinsically safe adaptive fault-tolerant control method for an overdriven vector rotorcraft according to claim 1, characterized in that: f1, f2, f3, f4, f5, and f6 represent the vector thrusts of the six vector propulsion units respectively; According to the vector thrust f i , calculate the steering gear rotation angle α in the i-th vector propulsion unit i 、External servo rotation angle λ i and motor speed Among them, c f is the aerodynamic constant.

6. A computer-readable storage device storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for intrinsically safe adaptive fault-tolerant control of an overdriven vector rotorcraft as claimed in any one of claims 1 to 5 are implemented.

7. An intrinsically safe adaptive fault-tolerant control device for an overdriven vector rotorcraft, comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, characterized in that: The processor executes the computer program to implement the steps of the method for intrinsically safe adaptive fault-tolerant control of an overdriven vector rotorcraft according to any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for intrinsically safe adaptive fault-tolerant control of an overdriven vector rotorcraft as claimed in any one of claims 1 to 5 are implemented.

Images