Comparison of the passive dynamics of walking on ground, tied-belt and split-belt treadmills, and via the Gait Enhancing Mobile Shoe (GEMS)

Abstract

This research compares walking over ground, on a split-belt treadmill, on a tied-belt treadmill, and on the Gait Enhancing Mobile Shoe (GEMS) in both humans and simulated on a passive dynamic model. Passive Dynamic Walkers (PDW) have been researched for decades, yet only recently has the model been used significantly in gait rehabilitation. We aim to identify how well the two-dimensional PDW can be used as a kinematic approximation tool for gait analysis. In this work, the PDW was scaled according to an anthropomorphic human model. For comparison, measurements were taken of humans walking in the same four environments. For normal walking, the PDW was found to be a good approximation for symmetric and rhythmic hip position, foot position, and velocity profiles. Tied-belt and split-belt treadmill model estimations revealed that the PDW's lack of dorsiflexion, joint stiffness, and joint damping limited the comparison, however trends between the human and the model agreed. The kinematics of the GEMS showed good agreement in interlimb interactions indicating that the PDW can be used as a good kinematic predictor for the GEMS.