Effects of system nonlinearities on posture adjustments using functional neuromuscular stimulation

A real-time simulation of a musculoskeletal model has been implemented to facilitate the development of a functional neuromuscular stimulation (FNS) standing system that provides the user with the ability to make adjustments to posture. The system has been developed as a platform to investigate the effects of nonlinearities on posture adjustment capability and to explore strategies for using adaptive filters to improve system performance. The standing spinal cord injured subject is modeled as an inverted pendulum with a single degree of freedom. The simulation user operates a standard computer input device to differentially control the stimulation applied to a pair of muscles in order to balance the system model. Visual feedback is provided by a real-time display on the computer monitor. The system provides the option of simulating various muscle models ranging in complexity from instantaneous torque generators to models that include nonlinear recruitment, muscle activation dynamics, muscle activation delay, torque-angle, and torque-velocity properties. In addition, the system includes real-time nonlinear adaptive filters designed to linearize the input/output properties as seen by the user. This simulation system is currently being used in studies to develop and evaluate improved systems for FNS standing and maneuvering.