Bae, Joonbum (University of California at Berkeley), Tomizuka, Masayoshi (Univ of California, Berkeley)

Gait Phase Analysis Based on a Hidden Markov Model

Scheduled for presentation during the Regular Session "Robotics" (WeB1), Wednesday, September 15, 2010, 14:40−15:00, Salon 1

5th IFAC Symposium on Mechatronic Systems, September 13-15, 2010, Marriott Boston Cambridge, Cambridge, Massachusetts

This information is tentative and subject to change. Compiled on May 25, 2019

Abstract

For effective rehabilitation treatments, the status of a patient's gait needs to be analyzed precisely. Since the gait motions are cyclic with several gait phases, the gait motions can be analyzed by gait phases. In this paper, a hidden Markov model (HMM) is applied to analyze the gait phases in the gait motions. Smart Shoes are utilized to obtain the ground contact forces (GCFs) as observed data in the HMM. The posterior probabilities from the HMM are used to infer the gait phases, and the abnormal transition between gait phases are checked by the transition matrix. The proposed gait phase analysis methods are applied to actual gait data, and the results shows that the proposed methods can be used to diagnose the status of a patient and evaluate a rehabilitation treatment.