Format

Send to

Choose Destination
J Appl Physiol (1985). 2015 Sep 1;119(5):541-57. doi: 10.1152/japplphysiol.01133.2014. Epub 2015 Jul 9.

An experimental comparison of the relative benefits of work and torque assistance in ankle exoskeletons.

Author information

1
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania; and.
2
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania; and Robotics Institute, Carnegie Mellon University, Pittsburgh, Pennsylvania stevecollins@cmu.edu.

Abstract

Techniques proposed for assisting locomotion with exoskeletons have often included a combination of active work input and passive torque support, but the physiological effects of different assistance techniques remain unclear. We performed an experiment to study the independent effects of net exoskeleton work and average exoskeleton torque on human locomotion. Subjects wore a unilateral ankle exoskeleton and walked on a treadmill at 1.25 m·s(-1) while net exoskeleton work rate was systematically varied from -0.054 to 0.25 J·kg(-1)·s(-1), with constant (0.12 N·m·kg(-1)) average exoskeleton torque, and while average exoskeleton torque was systematically varied from approximately zero to 0.18 N·m·kg(-1), with approximately zero net exoskeleton work. We measured metabolic rate, center-of-mass mechanics, joint mechanics, and muscle activity. Both techniques reduced effort-related measures at the assisted ankle, but this form of work input reduced metabolic cost (-17% with maximum net work input) while this form of torque support increased metabolic cost (+13% with maximum average torque). Disparate effects on metabolic rate seem to be due to cascading effects on whole body coordination, particularly related to assisted ankle muscle dynamics and the effects of trailing ankle behavior on leading leg mechanics during double support. It would be difficult to predict these results using simple walking models without muscles or musculoskeletal models that assume fixed kinematics or kinetics. Data from this experiment can be used to improve predictive models of human neuromuscular adaptation and guide the design of assistive devices.

KEYWORDS:

ankle foot orthosis; biomechanics; gait; locomotion; rehabilitation

PMID:
26159764
DOI:
10.1152/japplphysiol.01133.2014
[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center