Gait recognition for exoskeleton robots based on improved KNN-DAGSVM fusion algorithm

摘要

Currently exoskeleton robots have a wide range of applications in medical bionics, and care projects for the elderly and disabled. The recognition accuracy of human gait and the real-time performance of the system remain to be improved with urgent need. The conventional KNN method and the DAGSVM algorithm for gait detection are used in this research to divide a whole gait cycle of a human walking on level ground into five stages. It proposes a joint fusion algorithm (im-proved KNN-DAGSVM algorithm) on the basis of KNN algorithm and DAGSVM algorithm. The results reveal that the im-proved KNN-DAGSVM algorithm can successfully improve the recognition rate while shortening identification time.