Gaussian process learning and interpolation of gait motion for rehabilitation robots

Changmook Chun, Seung-Jong Kim, Jisoo Hong, Frank C. Park

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

2 Citations (Scopus)

Abstract

We present an alternative approach to generate gait motion at arbitrary speed for gait rehabilitation robots. The methodology utilizes Gaussian process dynamical model (GPDM), which is a nonlinear dimensionality reduction technique. GPDM consists of a dynamics in low-dimensional latent space and a mapping from the space to configuration space, and GPDM learning results in the low-dimensional representation of training data and parameters for the dynamics and mapping. We use second-order Markov process dynamics model, and hence given a pair of initial points, the dynamics generates a latent trajectory at arbitrary speed. We use linear regression to obtain the initial points. Mapping from the latent to configuration spaces constructs trajectories of walking motion. We verify the algorithm with motion capture data from 50 healthy subjects, who walked on a treadmill at 1, 2, and 3 km/h. We show examples and compare the original and interpolated trajectories to prove the efficacy of the algorithm.

Original languageEnglish
Title of host publicationICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications
EditorsDonald Bailey, Serge Demidenko, G. Sen Gupta
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages198-203
Number of pages6
ISBN (Electronic)9781479964666
DOIs
Publication statusPublished - 2015 Apr 6
Externally publishedYes
Event6th International Conference on Automation, Robotics and Applications, ICARA 2015 - Queenstown, New Zealand
Duration: 2015 Feb 172015 Feb 19

Publication series

NameICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications

Conference

Conference6th International Conference on Automation, Robotics and Applications, ICARA 2015
CountryNew Zealand
CityQueenstown
Period15/2/1715/2/19

Fingerprint

Patient rehabilitation
Interpolation
Trajectories
Robots
Exercise equipment
Linear regression
Markov processes
Dynamic models
Data acquisition

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Cite this

Chun, C., Kim, S-J., Hong, J., & Park, F. C. (2015). Gaussian process learning and interpolation of gait motion for rehabilitation robots. In D. Bailey, S. Demidenko, & G. S. Gupta (Eds.), ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications (pp. 198-203). [7081147] (ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICARA.2015.7081147

Gaussian process learning and interpolation of gait motion for rehabilitation robots. / Chun, Changmook; Kim, Seung-Jong; Hong, Jisoo; Park, Frank C.

ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications. ed. / Donald Bailey; Serge Demidenko; G. Sen Gupta. Institute of Electrical and Electronics Engineers Inc., 2015. p. 198-203 7081147 (ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications).

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

Chun, C, Kim, S-J, Hong, J & Park, FC 2015, Gaussian process learning and interpolation of gait motion for rehabilitation robots. in D Bailey, S Demidenko & GS Gupta (eds), ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications., 7081147, ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications, Institute of Electrical and Electronics Engineers Inc., pp. 198-203, 6th International Conference on Automation, Robotics and Applications, ICARA 2015, Queenstown, New Zealand, 15/2/17. https://doi.org/10.1109/ICARA.2015.7081147
Chun C, Kim S-J, Hong J, Park FC. Gaussian process learning and interpolation of gait motion for rehabilitation robots. In Bailey D, Demidenko S, Gupta GS, editors, ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications. Institute of Electrical and Electronics Engineers Inc. 2015. p. 198-203. 7081147. (ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications). https://doi.org/10.1109/ICARA.2015.7081147
Chun, Changmook ; Kim, Seung-Jong ; Hong, Jisoo ; Park, Frank C. / Gaussian process learning and interpolation of gait motion for rehabilitation robots. ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications. editor / Donald Bailey ; Serge Demidenko ; G. Sen Gupta. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 198-203 (ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications).
@inproceedings{83877ad9c62e4023a8296e96f3f8a1e1,
title = "Gaussian process learning and interpolation of gait motion for rehabilitation robots",
abstract = "We present an alternative approach to generate gait motion at arbitrary speed for gait rehabilitation robots. The methodology utilizes Gaussian process dynamical model (GPDM), which is a nonlinear dimensionality reduction technique. GPDM consists of a dynamics in low-dimensional latent space and a mapping from the space to configuration space, and GPDM learning results in the low-dimensional representation of training data and parameters for the dynamics and mapping. We use second-order Markov process dynamics model, and hence given a pair of initial points, the dynamics generates a latent trajectory at arbitrary speed. We use linear regression to obtain the initial points. Mapping from the latent to configuration spaces constructs trajectories of walking motion. We verify the algorithm with motion capture data from 50 healthy subjects, who walked on a treadmill at 1, 2, and 3 km/h. We show examples and compare the original and interpolated trajectories to prove the efficacy of the algorithm.",
author = "Changmook Chun and Seung-Jong Kim and Jisoo Hong and Park, {Frank C.}",
year = "2015",
month = "4",
day = "6",
doi = "10.1109/ICARA.2015.7081147",
language = "English",
series = "ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "198--203",
editor = "Donald Bailey and Serge Demidenko and Gupta, {G. Sen}",
booktitle = "ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications",

}

TY - GEN

T1 - Gaussian process learning and interpolation of gait motion for rehabilitation robots

AU - Chun, Changmook

AU - Kim, Seung-Jong

AU - Hong, Jisoo

AU - Park, Frank C.

PY - 2015/4/6

Y1 - 2015/4/6

N2 - We present an alternative approach to generate gait motion at arbitrary speed for gait rehabilitation robots. The methodology utilizes Gaussian process dynamical model (GPDM), which is a nonlinear dimensionality reduction technique. GPDM consists of a dynamics in low-dimensional latent space and a mapping from the space to configuration space, and GPDM learning results in the low-dimensional representation of training data and parameters for the dynamics and mapping. We use second-order Markov process dynamics model, and hence given a pair of initial points, the dynamics generates a latent trajectory at arbitrary speed. We use linear regression to obtain the initial points. Mapping from the latent to configuration spaces constructs trajectories of walking motion. We verify the algorithm with motion capture data from 50 healthy subjects, who walked on a treadmill at 1, 2, and 3 km/h. We show examples and compare the original and interpolated trajectories to prove the efficacy of the algorithm.

AB - We present an alternative approach to generate gait motion at arbitrary speed for gait rehabilitation robots. The methodology utilizes Gaussian process dynamical model (GPDM), which is a nonlinear dimensionality reduction technique. GPDM consists of a dynamics in low-dimensional latent space and a mapping from the space to configuration space, and GPDM learning results in the low-dimensional representation of training data and parameters for the dynamics and mapping. We use second-order Markov process dynamics model, and hence given a pair of initial points, the dynamics generates a latent trajectory at arbitrary speed. We use linear regression to obtain the initial points. Mapping from the latent to configuration spaces constructs trajectories of walking motion. We verify the algorithm with motion capture data from 50 healthy subjects, who walked on a treadmill at 1, 2, and 3 km/h. We show examples and compare the original and interpolated trajectories to prove the efficacy of the algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84928782193&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84928782193&partnerID=8YFLogxK

U2 - 10.1109/ICARA.2015.7081147

DO - 10.1109/ICARA.2015.7081147

M3 - Conference contribution

AN - SCOPUS:84928782193

T3 - ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications

SP - 198

EP - 203

BT - ICARA 2015 - Proceedings of the 2015 6th International Conference on Automation, Robotics and Applications

A2 - Bailey, Donald

A2 - Demidenko, Serge

A2 - Gupta, G. Sen

PB - Institute of Electrical and Electronics Engineers Inc.

ER -