A GAIT PHASE CLASSIFICATION METHOD FOR STAIR WALKING USING FEATURE EXTRACTION AND MUSCLE SELECTION BASED ON EMG SIGNALS

摘要

The present invention relates to a gait phase classification method for stair walking based on electromyographic signals, comprising the following steps of: (a) receiving training signals of electromyographic signals of a plurality of muscles, which are measured during ascent walking and descent walking, to extract features from the electromyographic signals, and generating a shunt for classifying a gait phase of the walking with the extracted features by each of the features; (b) receiving test signals of the electromyographic signals of the muscles to test the shunt and calculate a recognition rate, selecting a shunt according to the recognition rate, and selecting a shunt for each of the ascent walking and the descent walking; and (c) recognizing a gait phase of the ascent walking or the descent walking by the selected shunt. According to the present invention, by using different features and muscles having a high recognition rate for each gait phase during each of the ascent walking and the descent walking, a gait phase for stair walking can be more accurately classified.