Learning force/motion control for constrained robot manipulators

This paper develops a motion/force learning control scheme for constrained robot manipulators. The proposed algorithm is able to improve the performance of constrained end-effectors based on the previous operation as the action is repeated, it is very simple and computationally efficient. A sufficient condition is given to ensure the convergence of both motion control and force control. The simulation results of a constrained manipulator are given to show the good performance.