A redundancy-based scheme to perform safe vision-based tasks amidst obstacles

This paper presents a redundancy-based scheme allowing to avoid both occlusions and obstacles for a mobile robot performing a vision-based task in a cluttered environment. We consider the model of a cart-like robot equipped with ultrasonic sensors and a camera mounted on a pan-platform. The proposed method relies on the continuous switch between several controllers depending on the environment. Experimental results validating this approach are given at the end of the paper.