Intelligent Vision-Based Part-Feeding on Dynamic Pursuit of Moving Objects

Abstract

The paper addresses the problem of picking up moving objects from a vibratory feeder with robotic hand-eye coordination. Since the dynamics of moving targets on the vibratory feeder are highly nonlinear and often impractical to model accurately, the problem has been formulated in the context of Prey Capture with the robot as a “pursuer” and a moving object as a passive “prey”. A vision-based intelligent controller has been developed and implemented in the Factory-of-the-Future Kitting Cell at Georgia Tech. The controller consists of two parts: The first part, based on the principle of fuzzy logic, guides the robot to search for an object of interest and then pursue it. The second part, an open-loop estimator built upon back-propagation neural network, predicts the target‘s position at which the robot executes the pickup task. The feasibility of the concept and the control strategies were verified by two experiments. The first experiment evaluated the performance of the fuzzy logic controller for following the highly nonlinear motion of a moving object. The second experiment demonstrated that the neural network provides a fairly accurate location estimation for part pick up once the target is within the vicinity of the gripper.