An Implementation of Neuro-Fuzzy System for Gait Analysis in a Smart Insole of Exoskeleton

Abstract

This study presents an implementation of neuro-fuzzy model in a smart insole for exoskeleton to extract features using for control. The features are found based on the walking state of the human. The model of fuzzy C means is applied based on the interval type 2 fuzzy. The gait phases of the smart insole are analyzed within a gait cycle of human motion. Due to the boundaries among the gait phases, fuzzy inference is used for finding these variations. In addition, the neural network structure uses as the training role for both the fuzzy membership function parameters and its weights. The results show that states of human motions can be filtered by the neuro-fuzzy model at 98.14%.