Date of Award

Spring 1997

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

H. Michael Lacker

Second Advisor

Peter Engler

Third Advisor

Stanley S. Reisman

Fourth Advisor

Timothy Nam Chang

Fifth Advisor

Judy Deutsch

Abstract

There exists hypothesis that gait selection is strongly correlated with mechanical energy efficiency in normal subjects. The hypothesis is experimentally proven, and intuitively taken for granted. However, it is not mathematically proven that the minimum energy consumption hypothesis is the underlying principle for the normal human gait. To prove the hypothesis we have developed a mathematical model of human walking, in which it is possible to predict an optimal gait at any given speed of walking based on the principle of minimum mechanical energy consumption.

This improved model, which includes the double-support phase of walking as well as the swing phase, is an extension to the previous model studied in the author's master thesis which included only the swing phase; with this improved model it is possible to calculate the mechanical energy loss during an entire walking cycle. This permits the unique determination of an optimal gait for any given speed of walking which minimizes the mechanical energy loss per unit length of motion. The hypothesis that minimum energy is consumed in normal gait is tested by comparing the predicted gait with that actually observed experimentally. Reasonable results are obtained and it is confirmed that minimum energy consumption is the underlying principle determining the characteristics of human gait. Nevertheless, there is some discrepancy between the theoretical and empirical data; to reduce the discrepancy it will be necessary to develop a more detailed model which permits, for example, the stance leg to bend, and the foot of the swing leg to move as an independent segment. To facilitate this task a generalized model of walking is developed and recommended for future research.

Share

COinS