Research On Fault Tolerant Control Of Manipulator Dynamics System With Actuator Faults

The robot manipulator is the most widely used automated mechanical equipment in the field of robotics,which can be used in industrial manufacturing,medical,military,semiconductor manufacturing and space exploration and other fields.Moreover,the continuous progress of manipulator technology has promoted the continuous expansion of its application field and scope,and the manipulator can complete many complex tasks well.However,long-term performance in extreme working environments and limited by the lack of its own mechanical structure,components such as actuators of the manipulator system will inevitably be abnormal,which will affect the control performance of the system and cause certain economic and property losses.How to quickly detect faults and eliminate the impact of faults after the manipulator fails has always been the main research content of fault diagnosis and fault-tolerant control of the manipulator.In this paper,the fault-tolerant control of manipulator system with actuator fault is studied.The main work is as follows:Firstly,aiming at the fault-tolerant control problems of actuator fault,external disturbance and system uncertainty in the manipulator system,a non-singular fast terminal sliding mode(NFTSM)controller fault-tolerant control method based on three-order sliding mode(TOSM)observer is designed.The third-order sliding mode observer is introduced to estimate both velocities of system and the lumped uncertain terms such as faults.A non-singular fast terminal sliding mode controller is designed by combining the estimation information of the lumped uncertainty term based on TOSM observer.The finite time stability and effectiveness of the designed fault tolerant controller are verified by theoretical analysis and comparative simulation.Secondly,a finite time fault-tolerant controller based on high gain observer and neural network is designed for the fault-tolerant control problem of robotic manipulator system with actuator fault,external disturbance,system uncertainty and unmeasurable system speed.By introducing a high-gain observer to estimate the system speed,using a radial basis function neural network(RBFNN)to approximate the upper bound of fault,uncertainty and external disturbance,combined with the estimation of the lumped uncertainty term provided by RBFNN,the NFTSM controller based on speed estimation and fault estimation is designed.Theoretical analysis and MATLAB simulation verify the effectiveness of the designed finite time fault-tolerant controller.Finally,aiming at the manipulator system under the influence of actuator faults and external disturbance,a fixed-time sliding mode fault-tolerant controller of the manipulator based on fixed-time second-order sliding mode observer is designed.By designing a fixed-time second-order sliding mode observer to estimate actuator faults and other lumped indeterminates.According to the total uncertainty terms such as the obtained fault information,a fixed-time sliding mode fault-tolerant controller is designed to compensate for the influence of the total uncertainty terms such as faults in the system.The simulation results verify the effectiveness and stability of the proposed second-order sliding mode observer and the fixed-time sliding mode fault-tolerant controller,respectively.