Location

6th Floor E14

Description

In this thesis Todd Farrell will present a novel algorithm that studies the role of extrinsic versus intrinsic sensing and determines a robust set of sensors from physical and reliability constraints for a robotic ankle prosthesis. Further, during this thesis a novel data-set that contains seven able-bodied subjects walking over 19 terrain transitions and seven non-amputee subjects walking over nine transitions constitutes the largest collection of transitions to date using an exhaustive set of sensors on the largest variety of terrain: inertial measurement units, gyroscopes, kinematics from motion capture, and electromyography from 16 sites on the lower limbs for non-amputee subjects and nine for amputee subjects. This work extends previous work by using more conditions, a larger subject group, and more sensors on amputees, and includes non-amputees.

The focus will be primarily on the application of the machine learning algorithm to sensor identification during rapid transitions in terrain from pre-foothold to just prior to post-foothold across different terrain boundaries. This work will advance the field of biomechatronics, our understanding of terrain adaptation in people both with and without amputations, contribute to the development of a fully terrain adaptive robotic ankle prosthesis, and improve the quality of life for the physically challenged.
Specifically we set out to prove between pre- and post-foothold the ankle and knee positions calculated using an IMU attached to an amputee’s powered prosthetic ankle can discriminate with greater than 99% accuracy between the following nine conditions:level ground and a 15 degree incline,

level ground to a 15 degree decline,

15 degree incline to 15 degree decline,

15 degree incline to level ground,

15 degree decline to a 15 degree incline,

15 degree decline to level ground

level ground to stair descent,

level ground to stair ascent,

level ground without incline or decline.

As a consequence, our results suggest that myography as a non-volitional sensing modality for terrain adaptive protheses will not be as practical as simply using purely intrinsic sensing.

Host/Chair: Hugh Herr

Participant(s)/Committee

Rosalind Picard, Joseph Paradiso