OPTIMIZED AUTOMATED TRACKING OF A MOVING OBJECT WITH A ROBOTIC EYE SYSTEM

This automated object tracking system is very essential in terms of surveillance system and many more applications. In this paper, the design of fuzzy proportional-integral-derivative controller is proposed to control a robot eye to achieve a real-time video object tracking. The controller uses vision-based object tracking as feedback and generates velocity commands for the robot eye to keep the object of interest (OBI) in video at the centre of the image frame. The particle filter (PF) is used to track OBI in video by estimating its position in image frame. A new optimized bootstrap PF (OBPF) is introduced solving problem of PF on video object tracking for rigid motion. A new adaptive OBPF (AOBPF) is introduced for tracking object that has variable speed. The experimental and comparison results of the designed controllers using different methods of optimization are presented to demonstrate and to verify the proposed video object tracking system performance.