Gait analysis of human locomotion based on motion capture system

This paper presents a segmentation method for human locomotion. The research on the rules of human locomotion plays an important role for the analysis of human movement strategy. We collected locomotion data from eighteen volunteers by motion capture system(MCS), and found that the measured stride length and frequency with speed had a mutation in the critical point between walking and running. To solve this problem, a segmentation method is proposed to describe the two gaits motion while the continuous formula is not suitable for using again. Especially, a running formula is proposed by introducing a compression of virtual leg based on the elastic linear inverted pendulum model(ELIPM), which can accurately describe the relationship between stride length, frequency and speed. By experimental comparisons, the proposed method can reduce 77% of the statistical error than Alexander´s one, which indicates this method is an effective approach and can uniform the motion rules for human walking and running.