Automatic program generation from teaching data for the hybrid control of robots

An efficient method is developed to generate programs for the hybrid position/force control of robots from teaching data. An operator's motion is measured in terms of the force exerted by the operator and the position of the end-effector. The acquired data are then analyzed in order to understand what the operator intended to do, and necessary information is obtained to generate the hybrid control program. Determinations are made of which control mode, position or force, should be taken in each direction, how much force should be exerted, and what trajectory the end-effector should follow. The interpreted motion is then translated into a robot language, which explicitly describes the motion strategy that the human operator conceived. The method was implemented on a direct-drive arm and a personal computer, and the efficiency of the method was demonstrated through experiments.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>