State Observer-Based Backstepping Terminal Sliding Mode Control for Rectangular Robot Systems

Abstract

This study focuses on the fast state estimation of unmeasured state variable and uncertainty by using super-twisting state observer and adaptive law, and the design of a backstepping-based terminal sliding mode controller for rectangular robot systems. Introducing a super-twisting state observer can estimate unmeasured velocity information more rapidly than a conventional high-gain state observer. A backstepping controller with super-twisting observer is combined with a terminal sliding mode control scheme, which demonstrates faster error convergence performance than conventional backstepping sliding mode control. This estimation law, combined with the robustness of the controller for unknown states and dynamics, results in outstanding control performance when compared to conventional model-based computed torque control methods. The stability of the proposed control system was verified by using the Lyapunov-candidate-function. Comparative simulation and experimental results for the two-axis rectangular robot system demonstrate the efficacy of the proposed control scheme.