Using mini minimum jerk model for human activity classification in home-based monitoring

Mostafa Ghobadi, Jacob Sosnoff, Thenkurussi Kesavadas, Ehsan T. Esfahani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes a method for human activity classification in home based monitoring. The proposed approach is based on minimum jerk (MinJerk), a primary model for smooth path planning employed by human motor control in upper-extremity motion. Based on new evidences that show common control strategies in lower and upper extremity, MinJerk is adapted in our study to estimate the foot motion with fifth order polynomial functions. Experimental data are recorded during walking and going up and down the stairs using a single inertial measurement unit. Features of interest in this study are the optimized curve fitting coefficients. Using a structured support vector machine with radial basis function, an overall accuracy of 98.6% is achieved for activity classification. The proposed method is also capable of detecting the transitions between the movements with accuracy of 99.96%.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics
Subtitle of host publicationEnabling Technology Festival, ICORR 2015
EditorsDavid Braun, Haoyong Yu, Domenico Campolo
PublisherIEEE Computer Society
Pages909-912
Number of pages4
ISBN (Electronic)9781479918072
DOIs
StatePublished - Sep 28 2015
Event14th IEEE/RAS-EMBS International Conference on Rehabilitation Robotics, ICORR 2015 - Singapore, Singapore
Duration: Aug 11 2015Aug 14 2015

Publication series

NameIEEE International Conference on Rehabilitation Robotics
Volume2015-September
ISSN (Print)1945-7898
ISSN (Electronic)1945-7901

Other

Other14th IEEE/RAS-EMBS International Conference on Rehabilitation Robotics, ICORR 2015
CountrySingapore
CitySingapore
Period8/11/158/14/15

Fingerprint

Human Activities
Stairs
Units of measurement
Monitoring
Curve fitting
Motion planning
Upper Extremity
Walking
Support vector machines
Foot
Lower Extremity
Polynomials
Support Vector Machine

Keywords

  • Home-Based Monitoring
  • Human Activity Classification
  • Minimum Jerk
  • Motor Control Model

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Rehabilitation
  • Electrical and Electronic Engineering

Cite this

Ghobadi, M., Sosnoff, J., Kesavadas, T., & Esfahani, E. T. (2015). Using mini minimum jerk model for human activity classification in home-based monitoring. In D. Braun, H. Yu, & D. Campolo (Eds.), Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics: Enabling Technology Festival, ICORR 2015 (pp. 909-912). [7281319] (IEEE International Conference on Rehabilitation Robotics; Vol. 2015-September). IEEE Computer Society. https://doi.org/10.1109/ICORR.2015.7281319

Using mini minimum jerk model for human activity classification in home-based monitoring. / Ghobadi, Mostafa; Sosnoff, Jacob; Kesavadas, Thenkurussi; Esfahani, Ehsan T.

Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics: Enabling Technology Festival, ICORR 2015. ed. / David Braun; Haoyong Yu; Domenico Campolo. IEEE Computer Society, 2015. p. 909-912 7281319 (IEEE International Conference on Rehabilitation Robotics; Vol. 2015-September).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ghobadi, M, Sosnoff, J, Kesavadas, T & Esfahani, ET 2015, Using mini minimum jerk model for human activity classification in home-based monitoring. in D Braun, H Yu & D Campolo (eds), Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics: Enabling Technology Festival, ICORR 2015., 7281319, IEEE International Conference on Rehabilitation Robotics, vol. 2015-September, IEEE Computer Society, pp. 909-912, 14th IEEE/RAS-EMBS International Conference on Rehabilitation Robotics, ICORR 2015, Singapore, Singapore, 8/11/15. https://doi.org/10.1109/ICORR.2015.7281319
Ghobadi M, Sosnoff J, Kesavadas T, Esfahani ET. Using mini minimum jerk model for human activity classification in home-based monitoring. In Braun D, Yu H, Campolo D, editors, Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics: Enabling Technology Festival, ICORR 2015. IEEE Computer Society. 2015. p. 909-912. 7281319. (IEEE International Conference on Rehabilitation Robotics). https://doi.org/10.1109/ICORR.2015.7281319
Ghobadi, Mostafa ; Sosnoff, Jacob ; Kesavadas, Thenkurussi ; Esfahani, Ehsan T. / Using mini minimum jerk model for human activity classification in home-based monitoring. Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics: Enabling Technology Festival, ICORR 2015. editor / David Braun ; Haoyong Yu ; Domenico Campolo. IEEE Computer Society, 2015. pp. 909-912 (IEEE International Conference on Rehabilitation Robotics).
@inproceedings{fcd974f6679b4e869ceab97c754f9d78,
title = "Using mini minimum jerk model for human activity classification in home-based monitoring",
abstract = "This paper proposes a method for human activity classification in home based monitoring. The proposed approach is based on minimum jerk (MinJerk), a primary model for smooth path planning employed by human motor control in upper-extremity motion. Based on new evidences that show common control strategies in lower and upper extremity, MinJerk is adapted in our study to estimate the foot motion with fifth order polynomial functions. Experimental data are recorded during walking and going up and down the stairs using a single inertial measurement unit. Features of interest in this study are the optimized curve fitting coefficients. Using a structured support vector machine with radial basis function, an overall accuracy of 98.6{\%} is achieved for activity classification. The proposed method is also capable of detecting the transitions between the movements with accuracy of 99.96{\%}.",
keywords = "Home-Based Monitoring, Human Activity Classification, Minimum Jerk, Motor Control Model",
author = "Mostafa Ghobadi and Jacob Sosnoff and Thenkurussi Kesavadas and Esfahani, {Ehsan T.}",
year = "2015",
month = "9",
day = "28",
doi = "10.1109/ICORR.2015.7281319",
language = "English (US)",
series = "IEEE International Conference on Rehabilitation Robotics",
publisher = "IEEE Computer Society",
pages = "909--912",
editor = "David Braun and Haoyong Yu and Domenico Campolo",
booktitle = "Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics",

}

TY - GEN

T1 - Using mini minimum jerk model for human activity classification in home-based monitoring

AU - Ghobadi, Mostafa

AU - Sosnoff, Jacob

AU - Kesavadas, Thenkurussi

AU - Esfahani, Ehsan T.

PY - 2015/9/28

Y1 - 2015/9/28

N2 - This paper proposes a method for human activity classification in home based monitoring. The proposed approach is based on minimum jerk (MinJerk), a primary model for smooth path planning employed by human motor control in upper-extremity motion. Based on new evidences that show common control strategies in lower and upper extremity, MinJerk is adapted in our study to estimate the foot motion with fifth order polynomial functions. Experimental data are recorded during walking and going up and down the stairs using a single inertial measurement unit. Features of interest in this study are the optimized curve fitting coefficients. Using a structured support vector machine with radial basis function, an overall accuracy of 98.6% is achieved for activity classification. The proposed method is also capable of detecting the transitions between the movements with accuracy of 99.96%.

AB - This paper proposes a method for human activity classification in home based monitoring. The proposed approach is based on minimum jerk (MinJerk), a primary model for smooth path planning employed by human motor control in upper-extremity motion. Based on new evidences that show common control strategies in lower and upper extremity, MinJerk is adapted in our study to estimate the foot motion with fifth order polynomial functions. Experimental data are recorded during walking and going up and down the stairs using a single inertial measurement unit. Features of interest in this study are the optimized curve fitting coefficients. Using a structured support vector machine with radial basis function, an overall accuracy of 98.6% is achieved for activity classification. The proposed method is also capable of detecting the transitions between the movements with accuracy of 99.96%.

KW - Home-Based Monitoring

KW - Human Activity Classification

KW - Minimum Jerk

KW - Motor Control Model

UR - http://www.scopus.com/inward/record.url?scp=84946100372&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84946100372&partnerID=8YFLogxK

U2 - 10.1109/ICORR.2015.7281319

DO - 10.1109/ICORR.2015.7281319

M3 - Conference contribution

AN - SCOPUS:84946100372

T3 - IEEE International Conference on Rehabilitation Robotics

SP - 909

EP - 912

BT - Proceedings of the IEEE/RAS-EMBS International Conference on Rehabilitation Robotics

A2 - Braun, David

A2 - Yu, Haoyong

A2 - Campolo, Domenico

PB - IEEE Computer Society

ER -