Fuzzy Sliding Mode Controller for a Flexible Single-Link Robotic Manipulator

Abstract

Two robust non-linear controllers have been developed in this study to control the rigid and flexible motions of a single-link robotic manipulator. The controllers consist of a conventional sliding mode controller (CSMC) and a fuzzy sliding mode controller (FSMC). The effects of fuzzy-tuning some of the CSMC control parameters on the overall performance of the arm have been investigated in this study. Furthermore, the proposed FSMC, whose parameters are determined by fuzzy inference systems, has been designed herein based on two Lyapunov functions. The rationale is to considerably reduce the momentum of the system before entering the boundary layer neighboring the sliding surface. This will significantly attenuate the structural deformations of the arm. The digital simulations have demonstrated that the structural deformations, incurred by the beam at the onset of its movement, can be significantly reduced by fuzzy-tuning some of the control parameters. Furthermore, the results have illustrated the superiority of the FSMC over the CSMC in producing a less oscillatory and more accurate response of the angular displacement at the base joint, in damping out the unwanted vibrations of the beam, and in requiring significantly smaller control torques.