Automation of a wheelchair mounted robotic arm using computer vision interface

Abstract

Assistive robotic devices have great potential to improve the quality of life for individuals suffering with movement disorders. One such device is a robot-arm which helps people with upper body mobility to perform daily tasks. Manual control of robot arms can be challenging for wheelchair users with upper extremity disorders. This research presents an autonomous wheelchair mounted robotic arm built using a computer vision interface. The design utilizes a robotic arm with six degrees of freedom, an electric wheelchair, computer system and two vision sensors. One vision sensor detects the coarse position of the colored objects placed randomly on a shelf located in front of the wheelchair by using a computer vision algorithm. The other vision sensor provides fine localization by ensuring the object is correctly positioned in front of the gripper. The arm is then controlled automatically to pick up the object and return it to the user. Tests have been conducted by placing objects at different locations and the performance of the robotic arm is tabulated. An average task completion time of 37.52 seconds is achieved.