Dissertation/Thesis Abstract

An Application Programming Interface for Inertial Sensors Used in Physical Therapy and Injury Prevention
by Khibin, Volodymyr Yevgenovich, M.S., University of California, Davis, 2011, 90; 1507157
Abstract (Summary)

An API for inertial sensors prototyped in MATLAB is presented, targeting applications relating to physical therapy and injury prevention. The API provides facilities for easy data calibration, mote orientation calculation using rotation matrices, gravity calculation, and velocity and displacement calculation using the Verlet algorithm. Lateral and angular velocity drift is canceled through a novel approach of mote stability detection and recalculation of absolute mote orientation with the gravity vector obtained from the accelerometer. Frequency detection is performed by the API on any sensor signal using autocorrelation and a linear Kalman filter is also implemented to filter gyroscope readings. The test platform for evaluation was the Shimmer 2R wireless sensing mote collecting accelerometer and gyroscope readings at a rate of 256 Hz. Evaluation was done with drills relating to anterior cruciate ligament (ACL) injury: a jump height measurement test, hop count test, and a drop test with angle measurements on the knee. Results show a mean height measurement error of 3.27±0.5 cm (14.36%), hop counting error of 0.95 hops (1.82%), knee flexion error of 4.11±1 degrees (5.09%), knee inward bend error of 4.12±1 degrees (15.3%) and outward of 6.66±1 degrees (9.06%). Portability of the API to an embedded platform is discussed with reevaluation of the experiments with lower sampling frequencies; a sampling rate as low as 64 Hz still shows good accuracy.

Indexing (document details)
Advisor: Ghiasi, Soheil
Commitee: Baas, Bevan, Wilken, Kent
School: University of California, Davis
Department: Electrical and Computer Engineering
School Location: United States -- California
Source: MAI 50/05M, Masters Abstracts International
Subjects: Computer Engineering, Electrical engineering, Computer science
Keywords: Accelerometer, Acl, Api, Gyroscope, Inertial sensor, Mote
Publication Number: 1507157
ISBN: 978-1-267-23899-3
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy