NARX muscle model based on Functional Electrical Stimulation (FES) using polynomial estimators and Singular Value Decomposition (SVD)

This paper presents the development of muscle model based on FES stimulation parameters using the Nonlinear Auto-Regressive model with Exogenous Inputs (NARX). FES stimulations with varying frequency, pulse width and pulse duration were used to estimate the muscle torque. The proposed approach managed to approximate the behavior of the system well with unbiased residuals.