Finite-Time Neural Network Fault-Tolerant Control for Robotic Manipulators under Multiple Constraints

In this study, a backstepping-based fault-tolerant controller for robotic manipulator system with input and output constraints was developed. First, barrier Lyapunov function adopted to ensure that the satisfied time-varying constraints. Subsequently, actuator saturation asymmetric dead-zone characteristics were also considered, described using continuous function. The impacts of failures unknown dynamical parameters eliminated by employing Gaussian radial basis neural networks. external disturbances compensated for, disturbance observer. Meanwhile, finite-time dynamic surface technique accelerate convergence errors. Finally, simulation 2-degrees-of-freedom showed effectiveness proposed controller.