A VOR-based vision tracking system using fuzzy logic control

This paper presents a vision tracking system which turns a camera towards a selected target under dynamic situations to achieve high recognition performance of a mobile robot. This feature is realized by a fuzzy logic controller calculating the camera turning angles to counteract the robot motion by estimating the motion of a mobile robot with information obtained from MEMS inertial sensors, wheel encoders, and a vision system. The controller can also detect efficiently any slips occurred while a robot moves. The proposed vision tracking system mainly consists of a sensor module and a pan motioned driving module with a camera installed. The proposed system is implemented on a mobile robot and evaluated experimentally, which shows that the developed system achieved the excellent tracking and recognition performance under various scenarios, especially during slip or collision occurrences under dynamic circumstances.