PD-sliding mode controller for robot manipulators: A comparison analysis

Abstract

This paper introduces a new robust position variable structure controller design method for robot position systems with parameter perturbations. In most situations, the exact knowledge of the robot dynamics does not exist. For this reason a robust variable structure PD-control scheme is proposed for the set-point regulation problem of robot systems. The controller compensates for the inexact information about the robot system. It uses only some robot bound parameters and measurable joint variables. The sufficient conditions are derived for the existence of a sliding mode in the robot system. The techniques of matrix norm inequalities are often addressed for robustness analysis of the controller. In addition, the stability conditions are also investigated in a global sense. Effective parameter-independent conditions are developed by using a full quadratic form of Lyapunov function. Comparative simulation results with classical PD-controller have been presented indicating that the sliding mode PD-like controller yields faster and smoother performance.