Complementary data fusion in vision-guide and control of robotic tracking

We present a data fusion control scheme for the hand-held camera of the SCO RBOT-ER VII robot arm for learning visual tracking and interception. The co ntrol scheme consists of two modules: The first one generates candidate act ions to drive the end-effector as accurate as possible directly above a mov ing target, so that the second module can handily take over to intercept it . The desired camera-joint coordinate mappings are generalized by Elman neu ral networks for a tracking module. The intercept module then determines a suitable intercept trajectory for the robot within the required conditions. The simulation results support the claim that it could be successfully app lied to track and intercept a moving target.