Fuzzy clustering of temporal parameters of gait during stance phase for walking speed estimation

Abstract

In this study we describe walking speed estimation method using only temporal parameters of gait. Further, this method can be adopted for automatic speed adaptation during walking on treadmill which is useful for gait rehabilitation process and also for navigation in virtual reality environment. The proposed speed estimation method is based on two temporal parameters during stance phase: time duration from heel contact (HC) to heel off (HO) and from heel contact (HC) to toe off (TO). To estimate walking speed, first, data for previously mentioned gait parameters are taken for several walking speed. After that, fuzzy clustering approach is used to identify how those two gait parameters affect walking speed. By using result from fuzzy clustering, subject's walking speed at any speed can be estimated by computing its membership. This method is inexpensive and easy to implement since it only uses simple foot switch.