Finite-Time Sliding Mode Control Based on Unknown System Dynamics Estimator for Nonlinear Robotic Systems

This brief proposes an unknown system dynamics estimator (USDE) based finite-time sliding mode control for nonlinear robotic systems with unknown dynamics. First, an USDE with only one tuned parameter is designed by using filter operations and invariant manifold to estimate the unknown system dynamics. Second, a novel sliding mode surface with faster response time is designed. By incorporating the USDE into the sliding mode surface, a composite sliding mode control (CSMC) is proposed to achieve better dynamic performance. With simple parameter turning, the CSMC can realize the high precision trajectory tracking control of robotic systems. Finally, simulation results illustrate the superiority of the proposed control strategy.