TARGET TRACKING BY CAR-LIKE MOBILE ROBOT USING VISUAL CONTROL

This work discusses the problem of mobile target tracking by a car-like robot, equipped with an embedded monocular vision system. A control strategy using features based tracking has been implemented in order to follow the mobile target, by keeping it centred in the image. In addition, the implemented technique keeps constant the distance between the robot and the target. By choosing simple shape target and integrating complementary algorithms, the tracking strategy becomes faster and more accurate. Experimental results are presented at the end to test the effectiveness of the visual tracking system.