Adaptive Fixed‐Time Fault‐Tolerant Control for Robotic Manipulators With Actuator Loss of Effectiveness Faults and Unknown Nonlinearities

Abstract

ABSTRACT In this study, an adaptive fixed‐time fault‐tolerant control method is proposed for ‐DOF robotic manipulators with actuator partial loss of effectiveness (LOE) faults and unknown nonlinearities. First, a novel adaptive law is designed to approximate the actuator LOE coefficients and compensate for the negative effects of the faults, which has the advantages of fast responses for sudden faults, and can be used to monitor the failure severity of the actuator. Meanwhile, the upper bounds of the unknown nonlinearities are estimated within a fixed time so that overestimated gains are not required to ensure system robustness, thereby avoiding excessive effort and obvious chattering. Then, an adaptive fault‐tolerant controller is developed to guarantee that each joint tracks the desired angle trajectory with high accuracy within a fixed time, regardless of the initial system states. The controller requires no prior knowledge about unknown nonlinearities and actuator faults, which facilitates practical implementation. Finally, the superiority of the proposed method is verified by comparative simulation and experimental results with other methods on a 4‐DOF manipulator under different scenarios.