Jaya Bhanu Rao Degala Ramalu
Newcastle University, UK
Title: Static and dynamic obstacle avoidance behaviour for powered wheelchair control system
Biography
Biography: Jaya Bhanu Rao Degala Ramalu
Abstract
Disabled and elderly people are to manoeuvre the powered wheelchair smoothly due to frail and lack in dexterity in handling joystick movement. Wheelchairs users require efficient transition in preventing and avoiding collisions to minimise fatigue and to reach destination safely. The collision cone approach has been proven to avoid static and dynamic obstacles successfully by researchers in mobile robots. This paper proposes an improvement to the static and dynamic obstacle avoidance behaviour for non-holonomic semi- autonomous powered intelligent wheelchair control system. The velocity obstacle approach (VO) is proposed because it is considered an easy and simple method for avoiding dynamic obstacles while collision cone approach is used to detect collision situation between two circular shaped objects. However, the challenge takes place in detecting non-circularshaped objects from sensor data. The collision avoidance behaviour has been further improved with integration of collision cone approach and fuzzy inference system. Experiments on various environmental scenarios such as partly cluttered areas have been demonstrated with the aid of player/stage simulation on Ubuntu12.04 LTS. A Saitek Cyborg Evo wireless joystick with 6 axes is used to drive the wheelchair in the player/stage simulation. The experimental results show significant improvement in detecting and avoiding collisions.
Speaker Presentations
Speaker PPTs Click Here