Vision Based Robot Navigation

The aspiration of this research work is to design vision based intelligent mobile robot navigation techniques to capacitate an autonomous robot to navigate in indoor/outdoor environments. The images from mobile robot, accoutered with a camera, are preprocessed and are used for drivable region extraction. A novel weighted matrix algorithm (WMA) is used for drivable region division for, both, constructed and unconstructed scenarios. Hurdle distance, reckoned from ultrasonic sensors information, is used in conjunction with extracted drivable road to generate the motion commands for road following and hurdle avoidance behavior. Navigation controller design has been actualized in MATLAB by exerting fuzzy logic soft computing techniques to engender the sought motion commands. The designed controllers are evaluated in a simulator for assorted hurdle and environment conditions prior to real time implementation on archetype robot.

Muhammad Amar (S’12) received the B.S. degree in electrical engineering from University of the Punjab Lahore, Pakistan, in 2010. He is currently working towards the Ph.D. degree from Monash University, Gippsland, VIC, Australia.His research interests include artificial neural networks, fuzzy logic, robotics and remote machine health monitoring.