Integrated visual servoing system to grasp industrial parts moving on conveyer by controlling 6DOF arm

Abstract

The paper describes an integrated visual servoing system able to track and grasp industrial parts moving on a conveyer by controlling a 6-DOF robot arm. Using a real-time image based visual servoing method, the fully-6-DOF robot arm is controlled by PC-based vision and robot controllers. The hybrid Kalman filter is incorporated into both controllers in order to track a moving object stably against visual data noise. A teaching method for grasping the moving objects is also proposed which leads to simplification of teaching tasks. Experimental results prove that the practical conveyer-tracking and grasping tasks in manufacturing can be realized with our integrated visual servoing system.