ANALYSIS AND IDENTIFICATION OF GAIT BALANCE STATES USING PLANTAR PRESSURE DISTRIBUTION

Abstract

 Among the various forms of accidents that occur in construction work, crashes and falls account for the largest percentage. In this study, we attempted to identify the balance state during walking from the measured plantar pressure data that were measured the changes in plantar pressure during balanced and unbalanced walking condition using plantar pressure sensors. Specifically, we proposed a method to convert plantar pressure data during linear walking under three different walking conditions into a feature vector during a walking cycle. Furthermore, we proposed a method to identify three types of walking conditions using a k nearest neighbor model on the transformed feature vectors. The identification results using our proposed method showed that balanced and unbalanced walking condition could be recognized with high accuracy.