Study of a class of T-S fuzzy-based Terminal Sliding Mode Control

Abstract

This study investigates hybrid design by using the Takagi-Sugeno(T-S) fuzzy system modeling method and the Terminal Sliding Mode Control (TSMC) technique. The combined scheme maintains the merits of both approaches. It can alleviate the on-line computational burden by using the T-S fuzzy model to approximate original nonlinear systems since most T-S fuzzy parameters can be computed off-line. Moreover, it can preserve the advantages of TSMC, including rapid response, robustness to disturbances and/or uncertainties, and especially the guarantee of the fast finite-time convergence after the state reaching a sliding surface. The proposed method is applied to a two-link robot manipulator dynamic system. It is also compared to three other methods: 1) Conventional Sliding Mode Control (CSMC) for original nonlinear systems, 2) TSMC for original nonlinear systems, and 3) the combination of the T-S fuzzy system and CSMC design. Simulation results show the benefits of the proposed scheme.