A METHOD OF GAIT PHASE PREDICTION USING LINEAR INTERPOLATION IN THE PROCESS OF GAIT PHASE RECOGNITION BY THE USER ADAPTIVE CLASSIFICATION BASED ON SEMG SIGNAL

摘要

The present invention relates to a method for predicting a walking step by using linear interpolation in a walking step recognition process of recognizing a walking step by using a classifier based on electromyography signals by using channel selection and adaptive characteristics to improve accuracy. The method comprises: (a5) a step of calculating an average required time for each walking step by using data for training of a walking step; (d1) a step of calculating a correction value by using the difference between an actual required time and a prediction required time for each walking step from the previous walking of the current walking to the current walking; (d2) a step of calculating a prediction correction time of a walking step of the current walking by using the correction value; and (d3) a step of predicting a required time of the walking step of the next walking by using the correction value, the average required time for each walking step, and the average change time from the previous walking of the current walking to the current walking. Through the method, the time for the walking step is more accurately predicted through the correction based on the linear interpolation, so it is possible to improve a recognition rate of the walking step, to more accurately recognize the walking step, and to more accurately control a dynamic will.