Evolving Fuzzy and Neural Network Models of Finger Dynamics for Prosthetic Hand Myoelectric-based Control

Abstract

This paper first presents a structure for prosthetic hand myoelectric-based control systems (MBCSs). A set of evolving Takagi-Sugeno fuzzy and neural network models of the human hand dynamics, i.e., the finger dynamics, is next offered. These models will be used as reference models in structures of MBCSs. The inputs of the models are the myoelectric signals obtained from eight sensors placed on human subject's arm, and the outputs of these models are the flexion percentages of midcarpal joint angles. Elements of model-based fuzzy control are included. This plenary keynote paper is supported by authors' recent papers on modeling in the framework of prosthetic hand myoelectric-based control.