ANALYSIS AND SIMULATION OF ADAPTIVE PD CONTROLLER FOR ROBOT MANIPULATORS

Abstract

Model-based adaptive PD 'controller is discussed for robot manipulators. This APD controller can be used for both tracking control and set point control. Unlike -most applications of adaptive control theory to robot manipulators, this analysis addresses the non-linear dynamics directly without approximation, linearization or ad-hoc assumptions. This APD control law utilizes a parameterization based on physical (time-invariant) quantities. 'the adaptive law is derived by using energy-like Lyapunov function which retains non-linear character and structure of dynamics, rather than simple quadratic Lyapunov functions which are popular in linear adaptive control systems.._ Further, this APD controller assures global asymptotical stability-of system. Referring to point to point control problem, the PD control law is made adaptive with respect to gravity parameters by estimating payload. On similar basis, the APD control law is derived for tracking control. In this basic PD control law is made adaptive with respect to inertia parameters, coriolis & centrifugal parameters & gravity parameters.