Multisegmental Sensor Integration for Balance Control

With this Phase I SBIR project we intend to develop technical solutions to the primary issues limiting the feasibility of a stand-alone real-time balance feedback device. We intend to develop a technique that uses multiple sensors to estimate the orientation of the user"s body segments in the presence of real world disturbances. The effort will develop and validate techniques for accurate real-time estimation of center of mass (COM) and center of pressure (COP). The methods will be implemented on an existing hardware platform, our lower extremity exoskeleton Ekso. We will leverage Ekso as a platform to increase development speed while still being in a human centered application. This effort will also seek to develop a balance classification method that can score the balance of a user. We intend to implement this balance classification method in a stationary standing posture but will design an extension to dynamic walking for use in a later phase of the project.