Recurrent Dynamic Neural Network Model for Myoelectric-based Control of a Prosthetic Hand

This paper discusses a novel nonlinear autoregressive network with exogenous inputs (NARX) neural network (NN) model of the nonlinear dynamic mechanisms occurring in the myoelectric-based control of prosthetic hand fingers. The experimental results demonstrate the good performance of the NN architecture in terms of output response and mean squared error. The comparison of experimental results also shows that the proposed NARX recurrent NN outperforms a linear recurrent NN with the same training algorithm.