Sliding mode fault-tolerant control method for aerial robots with unknown upper bounds of faults and disturbances

By designing adaptive control law and super torque non-singular fast terminal sliding mode surface in the aerial robot system, the problem of stability challenges of aerial robots under faults and disturbances is solved, and rapid and accurate fault and disturbance estimation and compensation are achieved, improving the The system is robust and stable, and is suitable for aerial robot systems with faults and disturbances with unknown upper boundaries.

CN118092495BActive Publication Date: 2024-11-22NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410101988.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-25
Publication Date
2024-11-22
Estimated Expiration
2044-01-25

AI Technical Summary

Technical Problem

Aerial robots face stability challenges brought by failures and disturbances when performing tasks, especially when the upper boundary information of the disturbance is unknown, it is difficult for the prior art to achieve rapid and accurate estimation and compensation of faults and disturbances, resulting in system stability and resilience. Not good enough.

Method used

A fault-tolerant control method for slip mode of aerial robot under the unknown upper boundary of faults and disturbances is designed, using adaptive control law and super torque non-singular fast terminal sliding mode surface to achieve rapid estimation and compensation of faults and disturbances, and weaken vibration and complete the system's fault-tolerant control within a limited time.

Benefits of technology

Without obtaining the upper bound information of the disturbance, it realizes rapid and accurate estimation and compensation of faults and disturbances, improves the robustness and stability of the system, shortens the convergence time, reduces the equipment resource requirements, and enhances the system's Application efficiency and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118092495B_ABST
    Figure CN118092495B_ABST
Patent Text Reader

Abstract

The present invention discloses a sliding mode fault-tolerant control method for an aerial robot with unknown upper bounds of faults and disturbances. Considering an aerial robot system with actuator faults and mechanical arm rotation disturbances with unknown upper bounds, a dynamic equation model is first established. Aiming at the singularity problem that cannot be avoided in traditional terminal sliding mode control methods, a non-singular fast terminal sliding mode control method is proposed to make the tracking error converge within a finite time. In order to suppress the common chattering problem of sliding mode control, an over-torque method is designed for the sliding surface. Since the disturbances caused by actuator faults and mechanical arm rotation are inevitable when performing tasks and their upper bounds are unknown, an adaptive control strategy is introduced to effectively estimate faults and disturbances. After combining with the over-torque non-singular fast terminal sliding mode control method, a sliding mode control law is designed. The present invention can be used for the fault-tolerant control implementation problem of an aerial robot system with actuator faults and mechanical arm rotation disturbances.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a sliding mode fault-tolerant control method for an aerial robot under unknown fault and disturbance upper bounds, and belongs to the technical field of fault-tolerant control of aerial robot systems. Background Art

[0002] With the continuous advancement of control technology and control accuracy, the safety of multi-rotor drones has been greatly improved, and they are widely used in various fields such as surface crack detection, aerial photography, disaster relief, etc. As demand continues to increase, the ability of multi-rotor drones to observe the environment can no longer meet people's needs. Therefore, multi-rotor drones also need to have the ability to directly interact with the external physical environment and transform the environment. In order to add this function to drones, people often install one or more robotic arms on drones, and this type of multi-rotor aircraft with robotic arms is usually called an aerial robot. Aerial robots with the ability to interact with the environment greatly increase the functionality of performing tasks in the air.

[0003] However, the related actions of the manipulator will also have a certain impact on the stable operation of the aircraft. When the aerial robot performs tasks that directly interact with the environment, it will not only face the risk of internal actuator failure, but also be disturbed by the rotation of the manipulator when performing aerial tasks. Because of this interference between the aircraft and the manipulator, the related control of the aerial robot faces greater challenges, and it has also attracted many researchers to conduct in-depth exploration. The aerial robot system is a highly coupled and under-driven system. The uncertainty, faults and interference in the system will have a serious impact on the stability of the system. In actual applications, uncertainty and external interference problems cannot be avoided, and the control method often has some jitter problems. How to ensure the stability of the aerial robot or even continue to perform aerial tasks when the actuator fails, while suppressing the jitter phenomenon in fault-tolerant control, and designing a more effective controller to improve the feasibility of the system has always been the focus of research by researchers.

[0004] In a large number of current research results, the upper bound information of the disturbance suffered by the system is often taken as a known quantity, which is unknown in actual engineering applications. To address this problem, a controller is designed that can quickly and accurately estimate and compensate for fault and disturbance information when the upper bound of the disturbance is unknown, which can improve the feasibility of the system in practical applications. Summary of the invention

[0005] Purpose of the invention: In view of the above research background, a sliding mode fault-tolerant control method for aerial robots with unknown upper bounds of faults and disturbances is proposed. An adaptive control law is designed to achieve rapid estimation and compensation of faults and disturbances without obtaining information about their upper bounds; a non-singular terminal sliding mode surface is designed to improve the robustness of the system. Compared with conventional terminal sliding mode control, this control law solves the singularity problem and can converge within a finite time at a faster speed; a super-torque method is designed to weaken the common jitter problem in sliding mode control. The designed controller can complete the fault-tolerant control of the aerial robot system within a finite time.

[0006] Technical solution:

[0007] The sliding mode fault-tolerant control method for an aerial robot with unknown upper bounds of faults and disturbances includes the following specific steps:

[0008] Step 1): Determine the dynamic model of the aerial robot;

[0009] Step 2): Determine the dynamic model of the aerial robot with actuator failure and manipulator rotation disturbance with unknown upper bounds;

[0010] Step 3): construct a super torque non-singular fast terminal sliding mode;

[0011] Step 4): Design an adaptive control law for the aerial robot. Based on Lyapunov stability theory, prove that the aerial robot can achieve fault-tolerant control under the action of the designed adaptive control law when facing actuator failures and manipulator rotation disturbances with unknown upper bounds.

[0012] Preferably, the specific implementation process of step 1) is:

[0013] Step 1.1): Establish the earth coordinate system OXYZ and the body coordinate system O V X V Y V Z V , the earth coordinate system OXYZ and the body coordinate system O V X V Y V Z V The conversion relationship is shown in (1):

[0014]

[0015] R represents the body coordinate system O V X V Y V Z V Transformation matrix to the earth coordinate system OXYZ:

[0016]

[0017] Among them, φ represents the roll angle of the aerial robot, θ represents the pitch angle of the aerial robot, and ψ represents the yaw angle of the aerial robot;

[0018] Step 1.2): Define the control input of the aerial robot as shown in (2):

[0019]

[0020] Where l represents the length from the rotor center to the center of gravity of the aerial robot, L i (i=1,2,3,4) represents the lift provided by each rotor of the aerial robot, c is the conversion coefficient between force and torque, u0,u4,u5,u6 are the control inputs of the aerial robot, u0=T1+T2+T3+T4, u4=l(T4-T2) / I x , u5=l(T3-T1) / I y ,u ψ =(Ω4-Ω3+Ω2-Ω1) / I z , where T i (i=1,2,3,4) represents the lift provided by each rotor, Ω i (i=1,2,3,4) represents the torque generated by each propeller, I x ,I y ,I z Represents the moment of inertia around the body coordinate axis, and the parameter i represents the index value;

[0021] Step 1.3): By using the Newton-Euler method and the Euler-Lagrange equation, the dynamic model of the aerial robot is obtained as follows:

[0022]

[0023] Where m is the mass of the aerial robot, (x, y, z) is the coordinate of the aerial robot's body center of gravity in the earth coordinate system OXYZ, g is the gravitational acceleration, and u1, u2, and u3 represent the virtual inputs of the aerial robot. The specific expressions are as follows:

[0024]

[0025] Preferably, the specific implementation process of step 2) is:

[0026] Step 2.1): Determine the dynamic model of the aerial robot with unknown upper bounds of actuator failure and manipulator rotation disturbance as shown in (5):

[0027]

[0028] Among them, f i(i=1,…,6) is an actuator fault with an unknown upper bound, d i (i=1,…,6) is the manipulator rotation perturbation with unknown upper bound, and parameter i represents the index value;

[0029] Step 2.2): For actuator fault f i and the manipulator rotation disturbance d i Make assumptions:

[0030]

[0031] Among them, F i , D i (i=1,…,6) are all unknown positive constants, |·| represents the absolute value, and the parameter i represents the index value.

[0032] Preferably, the specific implementation process of step 3) is:

[0033] Step 3.1): First define the aerial robot tracking error e i (i=1,…,6) as shown in (7):

[0034]

[0035] Among them, [x d y d z d φ d θ d ψ d ] is the expected value to be tracked, and parameter i represents the index value;

[0036] Step 3.2): Define the non-singular fast terminal sliding surface s ni As shown in (8):

[0037]

[0038] Among them, λ i , μ i 、h i (i=1,…,6) are all positive constants, parameter i represents the index value, and sgn is the sign function;

[0039] Design of super torque sliding surface ti As shown in (9):

[0040]

[0041] Among them, k1 and k2 are both positive constants, and t represents time;

[0042] Step 3.3): Combine equation (8) with equation (9) to obtain the super torque non-singular fast terminal sliding mode surface si As shown in (10):

[0043]

[0044] Preferably, the specific implementation process of step 4) is:

[0045] Step 4.1): Considering the actuator failure f i and the manipulator rotation disturbance d i uncertainty, design actuator fault f i The estimated error and the manipulator rotation disturbance d i The estimated error As shown in (11):

[0046]

[0047] in, The actuator fault f i The manipulator rotation disturbance d i The estimated value of ; therefore, the derivative of the estimated error is:

[0048]

[0049] Step 4.2): Design the control law as shown in (13):

[0050]

[0051] Among them, α i , β i (i=1,…,6) are all non-negative constants, and the parameter i represents the index value;

[0052] Step 4.3): The adaptive control law designed for the aerial robot is as follows:

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059] in ρ i ,σ i ,Λ i ,Σi is the fault-tolerant controller parameter, sgn(·) is the sign function;

[0060] According to Lyapunov stability theory, it is proved that the aerial robot can achieve fault-tolerant control under the action of the designed adaptive control law when facing actuator failures and manipulator rotation disturbances with unknown upper bounds.

[0061] Beneficial effects: For aerial robot systems with unknown upper bounds of faults and disturbances, a sliding mode fault-tolerant control method for aerial robots with unknown upper bounds of faults and disturbances is designed. An adaptive control law is designed to achieve rapid estimation and compensation of faults and disturbances without obtaining information about their upper bounds; a non-singular terminal sliding mode surface is designed to improve the robustness of the system. Compared with conventional terminal sliding mode control, this control law solves the singularity problem and can converge within a finite time with a faster convergence speed; a super-torque method is designed to weaken the common jitter problem in sliding mode control. In general, the present invention has the following advantages:

[0062] 1) An adaptive control law is designed to achieve fast and accurate estimation of fault and disturbance information under the condition of unknown upper bound information. While saving equipment resources, it also ensures fast and accurate fault compensation and disturbance compensation when external information is less.

[0063] 2) Based on the conventional terminal sliding surface, a super-torque method is designed to effectively suppress the chattering effect caused by sliding mode control and further improve the robustness of the system.

[0064] 3) The super-torque non-singular terminal sliding mode fault-tolerant control method designed by the present invention can solve the fault-tolerant control implementation problem of an aerial robot system with unknown upper bounds of faults and disturbances. The method can ensure that the system completes rapid convergence within a limited time, improves the efficiency of the equipment to a certain extent, reduces costs, and enhances the robustness and stability of the system.

[0065] 4) The adaptive super-torque non-singular fast terminal sliding mode fault-tolerant control method proposed in the present invention has high accuracy, strong safety, high efficiency, and not harsh hardware requirements. It has certain application significance and can be widely used in the fault-tolerant control implementation problem of a class of aerial robot systems with unknown upper bounds of faults and disturbances. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a model diagram of an aerial robot;

[0067] Figure 2 is a flow chart of the method of the present invention;

[0068] Figure 3 It is a comparison chart of the tracking error curves of the two controllers for x;

[0069] Figure 4 It is a comparison of the tracking error curves of the two controllers for y;

[0070] Figure 5 It is a comparison of the tracking error curves of the two controllers for z;

[0071] Figure 6 This is a comparison of the tracking error curves of the two controllers for the roll angle;

[0072] Figure 7 This is a comparison of the tracking error curves of the two controllers for the pitch angle;

[0073] Figure 8 This is a comparison chart of the tracking error curves of the two controllers for the yaw angle. DETAILED DESCRIPTION

[0074] The present invention will be further explained below in conjunction with the accompanying drawings.

[0075] like Figure 2 As shown in the figure, an adaptive law is designed for the case of actuator failure and unknown robot arm rotation disturbance, which realizes the rapid estimation and compensation of faults and interference without obtaining their upper bound information. When designing the non-singular terminal sliding surface, the super torque method is introduced to address the chattering problem caused by sliding mode control, which weakens the chattering problem. Compared with conventional terminal sliding mode control, this control law solves the singularity problem and can converge within a finite time, with a faster convergence speed and higher robustness. An adaptive super-torsion non-singular fast terminal sliding mode fault-tolerant control method for an aerial robot system is designed, including the following specific steps:

[0076] Step 1) Determine the aerial robot system dynamics model, including the following steps:

[0077] Step 1.1) Establish the earth coordinate system OXYZ and the body coordinate system O V X V Y V Z V , the conversion relationship between the two is shown in (1):

[0078]

[0079] Among them, R represents the transformation matrix from the body coordinate system to the earth coordinate system. φ, θ, and ψ represent the roll angle, pitch angle, and yaw angle of the aerial robot, respectively.

[0080] Step 1.2) Define the control input of the system as shown in (2):

[0081]

[0082] Where l is the length from the rotor center to the center of gravity of the aircraft, L i (i=1,2,3,4) represents the lift provided by each rotor, and c is the conversion factor between force and torque;

[0083] Step 1.3) By using the Newton-Euler method and the Euler-Lagrange equation, the motion equation model of the aerial robot can be obtained as:

[0084]

[0085] Where m is the mass of the system, (x, y, z) is the coordinates of the center of gravity of the body in the earth coordinate system, g is the gravitational acceleration, I x ,I y ,I z represents the moment of inertia around the body coordinate axis, u1, u2, and u3 represent the virtual input of the system. The specific expression is as follows:

[0086]

[0087] Step 2) determining the dynamic model of the aerial robot with actuator failure and manipulator rotation disturbance with unknown upper bounds, including the following steps:

[0088] Step 2.1) Determine the dynamic model of the aerial robot with faults and disturbances as shown in (5):

[0089]

[0090] Among them, f i (i=1,…,6) is the actuator failure, d i (i=1,…,6) is the disturbance caused by the rotation of the robot arm;

[0091] Step 2.2) For fault f i and the uncertain disturbance d i Make reasonable assumptions:

[0092]

[0093] Among them, F i , D i (i=1,…,6) are all unknown positive constants;

[0094] Step 3) constructing a super torque non-singular fast terminal sliding mode, comprising the following steps:

[0095] Step 3.1) First define the system tracking error as shown in (7):

[0096]

[0097] Among them, [x d y d z d φ d θ d ψ d ] is the expected value of tracking;

[0098] Step 3.2) Define the non-singular fast terminal sliding surface as shown in (8):

[0099]

[0100] Among them, λ i , μ i 、h i (i=1,…,6) are all positive numbers;

[0101] The designed super torque sliding surface is shown in (9):

[0102]

[0103] Among them, k1 and k2 are both positive constants;

[0104] Step 3.3) Combine equation (8) with equation (9) to obtain the super torque non-singular fast terminal sliding mode surface as shown in (10):

[0105]

[0106] Step 4) Design a fault-tolerant controller, including the following steps:

[0107] Step 4.1) Considering the uncertainty of the upper bound of fault and disturbance, the design estimation error is shown as (11):

[0108]

[0109] Therefore, taking the derivative of the estimation error, we get:

[0110]

[0111] Step 4.2) Design the adaptive control law as shown in (13):

[0112]

[0113] Among them, α i , β i (i=1,…,6) are all non-negative constants;

[0114] The control law designed for the system in step 4.3) is as follows:

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121] in sgn(·) is a sign function. According to Lyapunov stability theory, it can be proved that the aerial robot system can achieve fault-tolerant control under the action of the designed control law when facing faults and disturbances with unknown upper bounds.

[0122] The effectiveness of the implementation plan is illustrated by a real case simulation below.

[0123] The flight control system of the Qball-X4 quadcopter developed by Canada's Quanser company is used as the application research object. The relevant physical parameters of the aircraft are as follows:

[0124] Table 1 Aircraft related physical parameters

[0125]

[0126] The controller related parameters are as follows:

[0127] Table 2 Controller parameters

[0128]

[0129] In order to illustrate the effectiveness and superiority of the fault-tolerant control method designed in the present invention in solving the fault-tolerant control problem of an aerial robot system with actuator failure and mechanical arm rotation interference, it is assumed that the initial state of the aircraft is [x0, y0, z0, φ0, θ0, ψ0] = [0, 0.2, 0.2, 0.1, 0.2, -0.1], and the aerial robot moves to the desired state [x d ,y d ,z d ,φ d ,θ d ,ψ d ]=[0.4,-0.4,0.6,0,0,0], there is always external interference, and the following faults occur in each control channel:

[0130] f i =0.3+0.1sint 6≤t≤9(i=1,…,6)

[0131] At the same time, the robotic arm rotates slightly at the 14th second, and the disturbance is as follows:

[0132] d i =0.1cost 14≤t≤17(i=1,…,6)

[0133] In order to better demonstrate that the adaptive super-torque non-singular terminal sliding mode fault-tolerant controller designed in the present invention has higher performance, a traditional terminal sliding mode controller is selected for comparison with it. Figure 3 , Figure 4 , Figure 5 It is a comparison chart of the tracking error curves of the two controllers for position; Figure 6 This is a comparison of the tracking error curves of the two controllers for the roll angle. Figure 7 This is a comparison of the tracking error curves of the two controllers for the pitch angle. Figure 8 This is a comparison of the tracking error curves of the two controllers for the yaw angle. It can be seen from the curves that both methods can achieve the fault-tolerant control goal of the aerial robot system under actuator failure and disturbance, but it is obvious that the controller designed by the present invention has a shorter convergence time, a faster convergence speed, a stronger suppression effect on the chattering phenomenon, and better robustness. Therefore, this case can show that the control method is effective.

[0134] The present invention designs an adaptive law for the situation of actuator failure and unknown mechanical arm rotation disturbance, and realizes the rapid estimation and compensation of faults and interference without obtaining their upper bound information. When designing the non-singular terminal sliding surface, the super torque algorithm is introduced to reduce the jitter problem caused by sliding mode control. Compared with conventional terminal sliding mode control, this control law solves the singularity problem, can converge within a limited time, converges faster, and has higher robustness.

[0135] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A sliding mode fault-tolerant control method for an aerial robot with unknown upper bounds of faults and disturbances, characterized by: The specific steps include: Step 1): Determine the dynamic model of the aerial robot; Step 2): Determine the dynamic model of the aerial robot with actuator failure and manipulator rotation disturbance with unknown upper bounds; Step 3): construct a super torque non-singular fast terminal sliding mode; Step 4): Design an adaptive control law for the aerial robot. Based on Lyapunov stability theory, prove that the aerial robot can achieve fault-tolerant control under the action of the designed adaptive control law when facing actuator failures and manipulator rotation disturbances with unknown upper bounds; The specific implementation process of step 1) is: Step 1.1): Establish the earth coordinate system OXYZ and the body coordinate system O V X V Y V Z V , the earth coordinate system OXYZ and the body coordinate system O V X V Y V Z V The conversion relationship is shown in (1): R represents the body coordinate system O V X V Y V Z V Transformation matrix to the earth coordinate system OXYZ: Among them, φ represents the roll angle of the aerial robot, θ represents the pitch angle of the aerial robot, and ψ represents the yaw angle of the aerial robot; Step 1.2): Define the control input of the aerial robot as shown in (2): Where l represents the length from the center of the rotor to the center of gravity of the aerial robot, L i (i=1,2,3,4) represents the lift provided by each rotor of the aerial robot, c is the conversion coefficient between force and torque, u0,u4,u5,u6 are the control inputs of the aerial robot, u0=T1+T2+T3+T4, u4=l(T4-T2) / I x , u5=l(T3-T1) / I y ,u ψ =(Ω4-Ω3+Ω2-Ω1) / I z , where T i (i=1,2,3,4) represents the lift provided by each rotor, Ω i (i=1,2,3,4) represents the torque generated by each propeller, I x ,I y ,I z Represents the moment of inertia around the body coordinate axis, and the parameter i represents the index value; Step 1.3): By using the Newton-Euler method and the Euler-Lagrange equation, the dynamic model of the aerial robot is obtained as follows: Where m is the mass of the aerial robot, (x, y, z) is the coordinate of the aerial robot's body center of gravity in the earth coordinate system OXYZ, g is the gravitational acceleration, and u1, u2, and u3 represent the virtual inputs of the aerial robot. The specific expressions are as follows: The specific implementation process of step 2) is: Step 2.1): Determine the dynamic model of the aerial robot with unknown upper bounds of actuator failure and manipulator rotation disturbance as shown in (5): Among them, f i (i=1,…,6) is an actuator fault with an unknown upper bound, d i (i=1,…,6) is the manipulator rotation perturbation with unknown upper bound, and parameter i represents the index value; Step 2.2): For actuator fault f i and the manipulator rotation disturbance d i Make assumptions: Among them, F i , D i (i=1,…,6) are all unknown positive constants, |·| represents the absolute value, and parameter i represents the index value; The specific implementation process of step 3) is: Step 3.1): First define the aerial robot tracking error e i (i=1,…,6) as shown in (7): Among them, [x d y d z d φ d θ d ψ d ] is the expected value to be tracked, and parameter i represents the index value; Step 3.2): Define the non-singular fast terminal sliding surface s ni As shown in (8): Among them, λ i , μ i 、h i (i=1,…,6) are all positive constants, parameter i represents the index value, and sgn is the sign function; Design of super torque sliding surface ti As shown in (9): Among them, k1 and k2 are both positive constants, and t represents time; Step 3.3): Combine equation (8) with equation (9) to obtain the super torque non-singular fast terminal sliding mode surface s i As shown in (10): 。 2. The sliding mode fault-tolerant control method for an aerial robot with unknown upper bounds of faults and disturbances as claimed in claim 1, characterized in that: The specific implementation process of step 4) is: Step 4.1): Considering the actuator failure f i and the manipulator rotation disturbance d i uncertainty, design actuator fault f i The estimated error and the manipulator rotation disturbance d i The estimated error As shown in (11): in, The actuator fault f i The manipulator rotation disturbance d i The estimated value of ; therefore, the derivative of the estimated error is: Step 4.2): Design the control law as shown in (13): Among them, α i , β i (i=1,…,6) are all non-negative constants, and the parameter i represents the index value; Step 4.3): The adaptive control law designed for the aerial robot is as follows: where ρ i >0,σ i >0,Λ i ∈(0,1),Σ i ∈(0,1), ρ i ,σ i ,Λ i ,Σ i is the fault-tolerant controller parameter, sgn(·) is the sign function; According to Lyapunov stability theory, it is proved that the aerial robot can achieve fault-tolerant control under the action of the designed adaptive control law when facing actuator failures and manipulator rotation disturbances with unknown upper bounds.