Design and implementation of a vision based intelligent object follower robot

Abstract

In autonomous robot navigation system, integration of visual sensing sensor plays a vital role. In this paper a vision based differential drive object follower robot has been presented. The robot has been designed to work without any prior environmental information and without any GPS service. The object to be followed and position of the object is determined by processing a real time image feed by a camera. For stability in navigation of the robot while following object a hybrid automata model has been designed. Kalman filter has been used for tracking the object correctly.