Mobile Robot Path Planning Using Vision and Olfaction to Search for a Gas Source

Abstract

A new approach to search for a gas/odor source using a mobile robot is presented. When the presence of gas is detected, the robot turns in the direction of the airflow that carries the gas, and looks for any suspicious object. Path planning is performed using a potential field method. Unlike the classical obstacle avoidance techniques, a Mexican-hat shaped potential field is formed around the visually detected object. Consequently, the robot is guided along the perimeter of the object to sniff out the location of the gas leak. Experimental results are presented to demonstrate the validity of the proposed algorithm.