Intelligent active force control of a rigid robot arm using embedded iterative learning algorithm

Abstract

The paper presents a novel approach to estimating the inertia matrix of a robot arm adaptively and on-line using an iterative learning algorithm. It is employed in conjunction with an active force control strategy which has been shown to be very effective in accommodating the disturbances. A comprehensive study is performed on a rigid two link manipulator subject to a number of loading conditions. Results clearly indicate the effectiveness of the control scheme in compensating the disturbances and at the same time the estimated inertia matrix is optimized to values corresponding to the converged track error as learning progresses. The viability of the proposed control scheme is illustrated through an experimental work carried out on a robot arm.