Elsevier

Gait & Posture

Volume 57, Supplement 1, September 2017, Pages 356-357
Gait & Posture

The Utility of Passive Dynamic Walkers as a Proxy for Humans in Dynamic Balance Studies

https://doi.org/10.1016/j.gaitpost.2017.06.465Get rights and content

Introduction

Passive dynamic mechanical walkers (PDMWs) have been shown to be proxies for the study of human gait [1], [2]. Both humans and walkers can be modelled as inverted pendulums wherein the center of mass (COM) pivots passively over the foot. Previous studies have largely examined similarities in energy efficiency and stability based on Lyapunov exponents computed on various gait parameters [1], [2]. This study examines the similarities between human and a PDMW using normalized deviation of the COM, DN. DN measures the excursion of the COM in a coordinate system defined by the line joining the feet (IFL) which is the approximate center of the base of support during tripping and slipping [3]. Since balance loss via tripping or slipping induces a rotation of the subject, moment of inertia (MOI) considerations also play a role. Due to safety limitations required with human subjects in slipping and tripping trials, PDMWs may provide a useful way to study balance safely.

Section snippets

Research Question

Are measures of dynamic balance in PDMWs including MOI and DN similar enough to humans that PDMW can be used in place of humans in balance studies?

Methods

A wooden PDMW was built and marked with 10 retroreflective markers for three dimensional motion capture. The walker was filmed using an eight camera Motion Analysis Corporation Eagle system at 120 Hz. At the start of a trial, the walker was rocked gently to induce it to begin walking down a ramp inclined at 30 degrees. Nine trials were collected comprising 32 complete gait cycles. COM position, DN, and MOI about the IFL and the axes of each foot were calculated for the walker. Ensemble moment

Results

Mean DN curves for both the walker and the human subjects showed similar shapes during stable gait with no significant difference found where the error bars overlapped (right). Minimum DN occurred during heel strike, and maximum was during mid-swing. For both TD subjects and for the walker, the MOI about the IFL was smallest (left). However, the MOI data was otherwise dissimilar between walkers and humans.

Discussion

The two most common modes of falling involve tripping and slipping, both of which involve rotation. We have considered three likely axes about which falls can occur, the IFL, and the long axes of the feet (LAF). For lateral falls, one would expect rotation about the LAF. However, the MOI about these is larger than that about the IFL for both humans and for the walker implying that the IFL is the preferred axis of rotation. The double humped structure of the MOI curves is mirrored in the shape

References (4)

  • Yunping

    Mathematical Problems in Engineering

    (2013)
  • Collins

    Science

    (2005)
There are more references available in the full text version of this article.

Cited by (0)

View full text