Stereo Vision based Object Tracking Control for a Movable Robot Head

摘要

In this paper, we have developed a visual tracking control system on a purposely built movable robot head mounted on a Baxter robot. The tracking control algorithm used a stereo vision system provided by a Bumblebee camera, facilitated by the MATLAB computer vision toolbox. The objective of this work is to enable a robot to move its head as in a similar manner as our humans beings to focus vision on a moving object. Identification of object and estimation of its coordinate are performed based on image processing techniques. A fuzzy logic technique is applied to control the moving head in order to bring human like motion into the robot. The method developed in this work can be extended to any other movable robot head platform. Extensive experimental studies have been performed to test the effectiveness and efficiency of the proposed method.