INTELLIGENT VISION-BASED PART-FEEDING ON DYNAMIC PURSUIT OF MOVING-OBJECTS

The paper addresses the problem of picking up moving objects from a vi bratory feeder with robotic hand-eye coordination. Since the dynamics of moving targets on the vibratory feeder are highly nonlinear and oft en impractical to model accurately, the problem has been formulated in the context of Prey Capture with the robot as a ''pursuer'' and a mov ing object as a passive ''prey.'' A vision-based intelligent controlle r has been developed and implemented in the Factory-of-the-Future Kitt ing Cell at Georgia Tech. The controller consists of two parts: The fi rst part, based on the principle of fuzzy logic, guides the robot to s earch for an object of interest and then pursue it. The second part, a n open-loop estimator built upon back-propagation neural network, pred icts the target's position at which the robot executes the pickup task . The feasibility of the concept and the control strategies were verif ied by two experiments. The first experiment evaluated the performance of the fuzzy logic controller for following the highly nonlinear moti on of a moving object. The second experiment demonstrated that the neu ral network provides a fairly accurate location estimation for part pi ck up once the target is within the vicinity of the gripper.