Methods for Experimental Identification of Nonlinear Dynamic Systems of Unknown Form and Order With Application to Human Gait

University Of Wisconsin-Madison, Madison WI

Investigators

Abstract

This research seeks to develop new methods for interrogating experimental measurements of human gait in order to better understand how the neuromuscular system functions during this important process. A new nonlinear system identification strategy will be studied that utilizes a model for the system that is appropriate for small deviations from a nominal periodic motion (or average stride cycle). A frequency-domain output-only system identification method will be created to characterize the motion of the system about this periodic trajectory using linear time-periodic systems theory, and that will be used to find a mathematical model for the nonlinear system. A video-based motion capture system will be used to record the motion of a few subjects as they walk on an instrumented treadmill, and the measurements will be used to identify a reduced model for their neuromuscular function during gait. That model will then be incorporated into the computed muscle control forward dynamics algorithm in order to gain additional insights into the roles of individual muscles and the function of the neuromuscular system during gait. This work is expected to lead to improved treatments for gait disorders, potentially benefitting many of the thousands of people in the United States who currently suffer from these disorders. Common gait disorders may stem from birth defects, residual effects of a disease, such as polio, or from injuries sustained in athletic activities, accidents or military combat, and although numerous treatments have been proposed for gait disorders, it is currently very difficult to predict how effective treatment will be. This work is expected to increase our understanding of gait so that appropriate treatments can be chosen more reliably. Also, the framework that will be used in this work is general, so the methods developed can also be used to study other nonlinear systems such as vehicles, wind turbines, machinery, etc? A short course will be created based on the results of this research and it will be used to share these new methods with a broad audience.