Citation

Akiyama Y, Toda H, Ogura T, Okamoto S, Yamada Y. Gait Posture 2017; 60: 15-21.

Affiliation

Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi, Japan.

Copyright

(Copyright © 2017, Elsevier Publishing)

DOI

10.1016/j.gaitpost.2017.10.008

PMID

29128688

Abstract

The curving motion of the human body is more complex than gait motion for straight walking. In particular, when human can freely curve corners, the gait motion varies among and even within individuals. However, is it not possible to classify natural curving motion using a statistical method? This study investigates the natural curving motion, encompassing various walking paths selected by subjects, as opposed to previous studies that focused on specific stepping strategies or curving motion under precisely controlled conditions. As a result, the natural curving motions are statistically classified into five distinct groups based on certain motion indices. Each group represents a curving strategy and is mainly characterized by the inner inclination of the pelvis, outer rotation of hip joints at the time of heel contact of the inner leg, and inner and/or outer rotation of hip joints at the time of heel contact of the outer leg. Such strategies are speculated as typical motions within the large variation in natural curving motion. Another finding is that, unlike the joint pattern of lower limb joints in the sagittal plane, hip rotation and the abduction/adduction angle drastically change when curving. In particular, the large inner rotation and abduction angles of the hip joint of both legs, which reached approximately 30° and 10°, respectively, become important when considering the curving gait of a physical assistant robot. Our analysis and findings help specify the joint motion required for physical assistant robots.

Copyright © 2017 Elsevier B.V. All rights reserved.

Language: en

Keywords

Cluster analysis; Corner turn; Gait motion