Autonomous Wheelchair Control

Author:

Barnes, Matthew John

Area of Honors:

Mechanical Engineering

Degree:

Bachelor of Science

Document Type:

Thesis

Thesis Supervisors:

Dr. Sean N Brennan, Thesis Supervisor
Henry Joseph Sommer III, Thesis Honors Advisor
Dr. Hosam Kadry Fathy, Faculty Reader

Keywords:

Robotics
wheelchair
autonomous control

Abstract:

People with disabilities are increasingly reliant on technology to provide freedom of mobility. However, current technology requires extensive direct user joystick input, which limits functionality and creates error between desired and actual system output. This project aims to create an autonomous system for an electric wheelchair, resulting in improved freedom of mobility for people with disabilities. In an ideal system, the user's desired end location would be reached using advanced control algorithms, such that the device could implement all motor commands along the path. Conventional electric wheelchairs rely on users to manually direct the electric wheelchair through a series of cumbersome joystick maneuvers. In order to automate this process, the computer will need encoder feedback for motor control, LIDAR data for environment information, and path planning algorithms to achieve the desired end state. The project develops a robotic wheelchair testbed from the ground up with sensors, computers, and a power management system to conduct autonomous wheelchair experiments. Capabilities of the testbed are demonstrated with implementation of basic control algorithms, including closed-loop PID velocity and kinematic pose-following control laws.