Improving the realism of a wheelchair simulator through tracking of upper body movements

Abstract

Introduction Safe and efficient use of a manual wheelchair (MWC) requires appropriate training. New MWC users should learn how to maneuver in different environments (restricted spaces, uneven surfaces, etc.). Learning of appropriate balance skills is also crucial, to avoid falls when handling forward or side slopes. Moreover, long-term MWC use can lead to repetitive shoulder strain injuries, which can be avoided through adequate propulsion technique (increasing push amplitude, while decreasing cadence). We have developed a virtual reality MWC simulator for the training of MWC skills. The velocities of the rear wheels control a virtual MWC in a 3D scene presented on a computer screen in front of the participant, while force feedback simulates the effects of gravity and inertia. Our objective was to improve the realism of our existing wheelchair simulator, through the addition of upper body motion tracking that will provide feedback on trunk posture and propulsion movement. Material and methods Ten MWC users participated in the study. They propelled the virtual MWC while their trunk and arm movements were simultaneously recorded by three systems: – five inertial sensors placed on the trunk, arms and forearms; – a markerless, 3D depth camera system; – a high-precision motion capture system (gold standard). Upper body position as measured by the two low-cost approaches (inertial system and depth camera) was compared to the motion capture system. Results and conclusion Findings from this interdisciplinary study will optimize mobility of MWC users, through better MWC skills (balance and propulsion techniques), while possibly decreasing the incidence of falls and shoulder injuries.