Active Fault Tolerant Control of Uncertain Robotic System Based on Observer and Sliding Mode

Abstract

An active fault tolerant controller (AFTC) based on observer and sliding mode is proposed for uncertain robotic system with actuator fault. First, a third-order observer is developed to accurately detect and estimate the fault. Meanwhile, the uncertainties and velocity of the system are estimated, which can avoid the noise of velocity measurement. Then, an AFTC is obtained by dynamic fast terminal sliding mode (DFTSM) control with the estimated fault, uncertainties and velocity. Finally, a super-twisting sliding mode (STSM) algorithm is employed to reduce the chattering in the sliding mode control. The position tracking error is proved to converge to near zero in finite time via Lyapunov function analysis. A two-joint robot is used to verify the effectiveness of the proposed AFTC. Compared to the AFTC based on the second-order observer and conventional sliding surface, the proposed AFTC can ensure that the convergence rate of the position tracking error is faster and the accuracy is higher in finite time.