RESEARCH ON THE LAW AND QUANTIFICATION METHOD OF HUMAN LOWER LIMB GAIT SYNERGY BASED ON ADAPTIVE CPG

In this paper, we proposed a method based on the adaptive Hopf central pattern generator (CPG) to describe and measure the coordination laws of human lower limb movements during walking. Six young, healthy individuals (equal number of males and females) were asked to walk to a metronome beat while we collected angle data of their main joints during normal and two abnormal gaits. We used the Hopf-CPG network to characterize joint angles as amplitude and phase parameters. By examining the lower limb gait characteristics based on these parameters, we revealed the constraint relationship between lower-limb joints. We then used principal component analysis (PCA) to obtain a reduced dimensional description of the coordination characteristics of lower-limb gait, which suggested the existence of coordination characteristics shared by lower limbs during normal human walking at different step frequencies. Finally, we proposed a quantitative scoring method to measure gait coordination based on the coordination constraint law of lower limb joints, which yielded scores ranging from 0 to 1. Our results showed that the mean coordination scores for normal gait (greater than 0.93) were higher than those for toe walking (0.78) and knee-locked walking (0.51), which indicates that our method effectively quantifies lower limb coordination.