Assessment of Powered Ankle Exoskeleton on Human Stability and Balance

Abstract

While wearable robotic systems such as exoskeletons are designed to assist in human motion, they are often only studied in situations of level walking. In order to expand the types of actions for which exoskeletons can assist, the exoskeletons must be tested in a variety of situations to understand how users will respond to these systems. This thesis examines how a lower-limb exoskeleton (Dephy ExoBoot) in both actuated and unactuated states affect balance and stability when performing a balancing tasking on a beam. Data was collected via inertial measurement units and analyzed on a pooled level (with data from all subjects) and on an individual level. It was found that the exoskeleton in both states affects stride stability metrics (e.g., stride length, stride duration, and stride speed). Despite the changes in stride stability, balance overall (as measured by torso sway) remains unaffected by either exoskeleton state when considering the pooled subject data. This result indicates that the ExoBoot can be used in balance tasks without compromising the balance of the user. On an individual level, it was found that not all subjects followed these general trends, as each person moves in a unique manner. It was also found that subject who experienced the unpowered exoskeleton prior to the powered exoskeleton state developed stride strategies on the balance beam that were initially more conservative. Our findings suggest that lower-limb exoskeletons can be used for balancing tasks, and we recommend that balancing tasks should be included in the standards for exoskeleton evaluation.