Development of Assistive Technologies: Improvement of a Robotic Wheelchair Safety System and Integration of a Robotic Wheelchair with Biosensing Devices

Open Access
Baum, Taylor
Area of Honors:
Electrical Engineering
Bachelor of Science
Document Type:
Thesis Supervisors:
  • Dr. Sean Brennan, Thesis Supervisor
  • Dr. Tim Kane, Honors Advisor
  • assistive technology
  • robotic wheelchair
  • obstacle detection
  • motor neuron diseases
  • eyelid drive system
  • motion artifacts
  • brain-computer interfacing
Patients with Motor Neuron Diseases (MNDs) have minimal control over traditional electric wheelchairs. This reduced mobility severely impacts one's independence and quality of life. This work investigates the development of a robotic wheelchair platform for improved mobility for patients with MNDs. First, we improve the capabilities of a robotic wheelchair's safety system. This consists of an improved obstacle detection algorithm developed and then implemented on a robotic wheelchair platform. Using this algorithm, 7.5 cm negative obstacles could be detected with 90% accuracy with low-cost Light-Detection and Ranging (LiDAR) sensors. We also show a thorough characterization of the signals collected from said low-cost LiDAR sensors and potential pitfalls of the algorithm and sensors. Next, we explored the integration of a robotic wheelchair with multiple biosensing devices. We demonstrate the integration of a novel biosensing device, the Eyelid Drive System (EDS), with a robotic wheelchair and explored the impact of motion artifacts from a robotic wheelchair on the EDS signals. We found that the impact of a robotic wheelchair's motion varied significantly between different persons driving it and that these respective artifacts may be mitigated with machine learning algorithms paired with accelerometer data. A brief review of Brain-computer Interface (BCI) technology is presented, and future work will explore the integration of BCI technology with a robotic wheelchair. The work presented in this thesis may impact the quality of life of persons suffering from MNDs. We provide an obstacle detection algorithm, characterization of motion artifacts with the EDS system, and a general overview of BCI technologies. Each of these contributions may impact or influence future work in the development of robotic wheelchairs for use with patients with motor impairments.