Determination of Human Gait Phase Using Fuzzy Inference

Abstract

This paper discusses the design and implementation of a fuzzy inference system for the recognition of human gait phases. In particular, this work focuses on using the angles of the joints of lower limb to determine the current stage of a subject's gait cycle. The fuzzy rule-based system was developed using typical joint angle trajectories over a single gait cycle. The behavior of each joint was examined to determine appropriate rules for differentiating between gait phases. The completed system was then tested using joint angle trajectories measured from healthy human test subjects and shown to be capable of reproducing the gait phase transitions found by a human expert.