Robust adaptive fault‐tolerant control using RBF‐based neural network for a rigid‐flexible robotic system with unknown control direction
Haoping Wang, Xingyu Zhou, Yang Tian
Abstract
Abstract In this article, a radial basis function (RBF)‐based robust adaptive fault‐tolerant control (FTC) is developed for a class of rigid‐flexible robotic systems under actuator failures, unknown control direction, and uncertain external disturbances. The dynamics of the rigid‐flexible robotic over a vertical plane is governed by the hybrid ordinary differential equations–partial differential equations. The uncertain external disturbance is estimated via an RBF neural network. Meanwhile, the robust adaptive FTC law is utilized to follow the given joint angular and attenuate the vibration of the flexible structure in the case of actuator failures and unknown distributed disturbances. To handle the unknown control direction, the Nussbaum function is employed to robust FTC laws. By virtue of the Lyapunov direct approach, the trajectory tracking and vibration elimination for the controlled rigid‐flexible robotic system are proved and uniformly bounded in face of bounded external disturbances and unknown control directions. The efficacy of the developed control strategy is also illustrated via four comparison examples.