Real-time gait event recognition for wearable assistive device using an IMU on thigh

Songyot Piriyakulkit, Yasuhisa Hirata, Hiroshi Ozawa

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

2 Citations (Scopus)

Abstract

In this paper, we propose a real-time gait event recognition method to specifically use with a wearable device designed to assist patients with lumbar kyphosis. Our proposed method utilizes only a single IMU attached on the left thigh area of the wearable device. The gait events of interest in this paper are Toe-off(TO) and Heel-strike(HS) which indicates the start of Swing phase and Stance phase, respectively. We identify these gait events by the detection of local minima and maxima in angular velocity and linear acceleration profiles of a thigh with heuristic functions. The results show that the proposed method is viable for our wearable device.

Original languageEnglish
Title of host publication2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages314-318
Number of pages5
ISBN (Electronic)9781538631942
DOIs
Publication statusPublished - 2017 Jul 2
Event2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017 - Beijing, China
Duration: 2017 Oct 172017 Oct 19

Publication series

Name2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
Volume2018-January

Other

Other2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017
CountryChina
CityBeijing
Period17/10/1717/10/19

Keywords

  • gait event recognition
  • inertial measurement unit
  • wearable

ASJC Scopus subject areas

  • Artificial Intelligence
  • Control and Optimization

Fingerprint Dive into the research topics of 'Real-time gait event recognition for wearable assistive device using an IMU on thigh'. Together they form a unique fingerprint.

Cite this