A State-of-charge Estimator for A Semi-autonomous Electric Wheelchair

Open Access
Author:
Miller, Christopher Xavier
Area of Honors:
interdisciplinary in Electrical Engineering and Mechanical Engineering
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
  • Sean Brennan, Honors Advisor
  • Timothy Joseph Kane, Honors Advisor
  • Constantino Manuel Lagoa, Faculty Reader
  • Timothy Joseph Kane, Thesis Supervisor
Keywords:
  • wheelchair
  • state-of-charge
  • estimation
  • control
  • kalman filters
  • battery
  • robotics
  • circuits
  • assistive technology
Abstract:
According to the U.S. Census, nearly three million individuals in the United States rely on wheelchairs, a large portion of which are electric wheelchairs, in order to regain lost mobility. Consequently, these individuals depend on a reliable power system; if a wheelchairs battery power depletes without the user being aware, the individual may become stranded, further limiting his or her freedom of mobility and potentially placing the user in a harmful situation. This research seeks to develop a State-of-Charge (SOC) estimator for the batteries of an electric wheelchair. A second-order equivalent circuit battery model is developed and parametrized for a wheelchairs lead-acid battery pack. The inputs to the algorithm are battery voltage and current and the output of the algorithm is the battery packs estimated state of charge. To simplify the SOC estimation, this algorithm models a vehicles fuel gauge. When a vehicles fuel tank is nearly full or nearly empty, a fuel gauge presents the user with a full or empty reading. Outside of these regions, the fuel gauge varies directly with the fuel remaining in the vehicles tank. Similar to a vehicles fuel gauge, the algorithm yields the least accurate estimates of the wheelchairs SOC in the maximum and minimum SOC regions. These extrema are defined by the non-linearities present in the Open Circuit Voltage (OCV) SOC curve. Consequently, a coulomb accumulator is incorporated to estimate energy usage in these regions. A Kalman filter is incorporated to estimate SOC in the linear region of the OCV-SOC curve. This thesis presents the development of an autonomous wheelchair platform and the subsequent implementation of the aforementioned battery state estimator.