Neural Networks-Based Fault Tolerant Control of a Robot via Fast Terminal Sliding Mode

This article develops a robust fault tolerant (FT) control scheme for an n-link uncertain robotic system with actuator failures. In order to eliminate the influence of both the uncertainties and actuator failures on the system performance, the Gaussian radial basis function neural networks are used to compensate for the actuator failures and uncertain dynamics. An adaptive observer is designed to compensate for external disturbance. In addition, in order to accelerate the recovery of system stability after failure, a nonsingular fast terminal sliding mode is given. Finally, the simulation results on a two-link manipulator confirms the superior performance of the proposed neural networks-based FT controller, and the experiment results on the Baxter robot further verify the effectiveness of the control method.