Fuzzy Logic Based Implementation of a Real-Time Gait Phase Detection Algorithm Using Kinematical Parameters for Walking

Abstract

The concept of fuzzy logic was applied to develop a gait phase detection algorithm, to address the complexity of distinguishing between gait phases based on gait parameters obtained for walking. The proposed intelligent algorithm detects seven gait phases taking into consideration only joint parameters. Three inertial sensors were placed at the thigh, shank and foot in order to acquire hip, knee and ankle joint angles. The main objective is to incorporate the algorithm to a rehabilitation device in order to determine accurate timing for feedback. The gait phases detected could also be analyzed to identify normal and abnormal gait depending on the sequence of gait phases detected. Experiments were carried out to validate the feasibility of the algorithm with the acquisition of the joint parameters for five gait cycles. This paper also elaborates the results obtained along with the graphical representation of the gait parameters and the gait phases detected for normal and abnormal walking gait.