Robust Adaptive Fixed-time Trajectory Tracking Control of Manipulator based on Extreme Learning Machine

This paper mainly investigates the trajectory tracking control problems for manipulator systems with unknown dynamics and external disturbances. Firstly, an extreme learning machine (ELM) is adopt to compensate unknown dynamics of the manipulator. Then, an updating law is derived to ensure the convergence of ELM output wights. Besides, an indirect method is developed to avoid the potential singularity problem of the fixed-time sliding mode surface. Moreover, a robust adaptive controller is designed based on the outputs of ELM and sliding mode technique. By using the Lyapunov stability theory, the fixed-time convergence and stability of the adaptive control system can be guaranteed. Finally, simulation results is presented to show the efficiency of the proposed control structure with respect to different initial conditions.