An adaptive hybrid force and position learning control of robot manipulators

Abstract

An adaptive position/force learning control scheme is presented for the constrained robot motion. The hybrid position/force control strategy is formulated in the framework of learning control methodology and the developed learning controller converges to the desired inverse dynamics as learning continues. The properties of the learning system: its pointwise convergence, the rejection of unexpected disturbances in the tangential direction of the constrained surface are also presented.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>