Visual Servoing using Industrial Manipulator

Abstract

In this work, an image-based visual servoing technique is developed which uses the images from a RGB camera in eye-in-hand configuration with occlusion handling.The proposed method chooses a set of correctly extracted image features, and obtains an estimate of all the image features from the correctly extracted image features. The estimation procedure makes it possible to track image features even when occlusion occurs. A velocity controller is developed for a 6 -DoF robotic manipulator which minimises the error between current and desired imagefeatures. At each control loop, end-effector velocity is estimated from image interaction matrix and it is used to generate joint velocities from robot jacobian. An extensive set of experiments are carried out to indicate the feasibility of the proposed approach. Finally, a dexterous manipulation capability is shown by using the robot to play ping-pong.