Adaptive finite‐time fault‐tolerant control for flexible‐joint robotic stochastic systems with input saturation and sensor fault

Abstract

A new fault-tolerant approach to input saturation and sensor faults is presented for flexible-joint robotic stochastic nonlinear systems with finite-time convergence performance. More precisely, a new command filter is designed and embedded into the backstepping framework, which greatly reduces the amount of calculation. By using the estimation ability of fuzzy logic system and less adjustable parameters, the problems of sensor faults and unknown functions are solved, which facilitates the design of controllers. A smooth function and mean-value theorem are used to deal with the difficulty associated with system signals. A novel simpler controller is developed to ensure that the trajectory tracking error converges to a sufficiently small neighborhood around the origin within finite time, while there are random noises. Finally, the effectiveness of the presented scheme is verified by simulation results.

Reinforcement learning‐based adaptive predefined‐time optimal tracking control for strict‐feedback nonlinear systems

“…Step i (2 ≤ i ≤ n − 1): By utilizing the transformation specified in (6), the error dynamic equation can be expressed as…”

“…The topics concerning nonlinear control 1,2 have garnered considerable interest during the past decades. In the practical system dynamics, numerous nonlinear systems can be depicted as strict-feedback nonlinear structures, such as spacecraft attitude control systems, 3,4 robotic systems, 5,6 and unmanned aerial vehicle attitude systems. 7,8 One of the dominant control strategies employed in such systems is the backstepping technique.…”

Observer‐based finite‐time adaptive fault‐tolerant control for nonlinear system with input delay and prescribed performance

“…In recent years, fault‐tolerant control has received considerable attention. The fault‐tolerant control schemes have been proposed in References 4‐10. An adaptive output feedback fault‐tolerant control strategy for non‐affine nonlinear systems with unmeasurable states was proposed in Reference 4.…”

“…Based on the adaptive observer, a fast adaptive fault estimation algorithm for Markov jump systems was proposed by using the backstepping method in Reference 6. Recently, in References 7 and 8, the adaptive control schemes have emerged for stochastic nonlinear systems with actuator failures. In References 9 and 10, the actuator fault problem of nonlinear interconnected large‐scale systems was studied, and a decentralized control strategy was proposed to solve the actuator fault.…”