A Unified Adaptive Control Law Design for Rigid Body Robots via a Hamiltonian Property

Abstract

A unified adaptive control scheme for rigid body robots is presented which has, as special cases, the most recent control laws, such as variable structure control laws and adaptive control laws obtained by exploiting a Hamiltonian property of rigid body robots. The resulting control laws are free of troublesome inertia-matrix inversions. Integral error control is explicitly introduced to effectively eliminate static tracking errors. A separation between control design and parameter adaptation is obtained. Two general parameter adaptation algorithms are given based on the method of passivity, guaranteeing boundedness of all the signals.