Date of Award

1-1-1995

Document Type

Thesis

Publisher

Edith Cowan University

Degree Name

Master of Science

Faculty

Faculty of Science, Technology and Engineering

First Supervisor

Dr Zhihong Man

Abstract

This thesis investigates the problem of robust adaptive sliding mode control for nonlinear rigid robotic manipulators. A number of robustness and convergence results are presented for sliding mode control of robotic manipulators with bounded unknown disturbances, nonlinearities, dynamical couplings and parameter uncertainties. The highlights of the research work are summarized below : • A robust adaptive tracking control for rigid robotic manipulators is proposed. In this scheme, the parameters of the upper bound of system uncertainty are adaptively estimated. The controller estimates are then used as controller parameters to eliminate the effects of system uncertainty and guarantee asymptotic error convergence. • A decentralised adaptive sliding mode control scheme for rigid robotic manipulators is proposed. The known dynamics of the partially known robotic manipulator are separated out to perform linearization. A local feedback controller is then designed to stabilize each subsystem and an adaptive sliding mode compensator is used to handle the effects of uncertain system dynamics. The developed scheme guarantees that the effects of system dynamics are eliminated and that asymptotic error convergence is obtained with respect to the overall robotic control system. • A model reference adaptive control using the terminal sliding mode technique is proposed. A multivariable terminal sliding mode is defined for a model following control system for rigid robotic manipulators. A terminal sliding mode controller is then designed based on only a few uncertain system matrix bounds. The result is a simple and robust controller design that guarantees convergence of the output tracking error in a finite time on the terminal sliding mode.

Share

 
COinS