Development of an obstacle detection and navigation system for autonomous powered wheelchairs

This thesis describes the development and prototyping of an after market system to convert an electric powered wheelchair into an autonomous wheelchair. The purpose of this research is to automatize powered wheelchairs for children who are impacted by conditions such as cerebral palsy. Maneuvering a powered wheelchair with a joystick is difficult and painful for users who have a high level of cerebral deficiency and other chronic conditions. The autonomous powered wheelchair is designed to maneuver in an indoor environment whilst avoiding static and dynamic obstacles. The add-on system comprised of stereo vision sensors (ZED Camera), and IMU is designed to use Robot Operating System (ROS) to communicate and control the movement of the wheelchair. With the addition of a 3D map of the environment generated using visual sensors through ROS packages, the system identifies and avoids obstacles. Simultaneous Localization and Mapping (SLAM) and autonomous navigation packages were tested and modified. Slopes and drops identified in the 3D map are converted such that they are compatible with the 2D navigation packages of ROS. Configuration settings were determined and tested to ensure that the system works as required. The results demonstrated that the powered wheelchair can be modified to become an autonomous wheelchair using ZED Camera and IMU, such that it can navigate indoors effectively avoiding static and dynamic obstacles.

Name thesis:degree_name

Master of Applied Science (MASc)

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Ontario Institute of Technology

Year dc:date.issued

2019

Author dc:creator

• Diwan, Heta

Advisor dc:contributor.advisor

• Nokleby, Scott

• Robotics
• Autonomous wheelchair
• Obstacle detection
• Slope and stair detection
• Autonomous navigation

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10155/1051

OAI identifier oai:identifier

oai:ontariotechu.scholaris.ca:10155/1051