Ambulatory activity classification with dendogram-based support vector machine: Application in lower-limb active exoskeleton.
Ambulatory activity classification is an active area of research for controlling and monitoring state initiation, termination, and transition in mobility assistive devices such as lower-limb exoskeletons. State transition of lower-limb exoskeletons reported thus far are achieved mostly through the u...
| Publicado en: | Gait & Posture Vol. 50; pp. 53 - 60 |
|---|---|
| Autores principales: | , , , |
| Formato: | research Journal Article |
| Publicado: |
Elsevier B.V.
Oct2016
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=118965610&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 118965610 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 09666362 3L6 jtl: Gait & Posture issn: 09666362 maglogo: N pubinfo: dt: Oct2016 vid: 50 pid: 1004 pub: Elsevier B.V. artinfo: ui: 118965610 118965610 NLM27585182 118965610 10.1016/j.gaitpost.2016.08.010 NLM27585182 118965610 ppf: 53 ppct: 7 formats: tig: atl: Ambulatory activity classification with dendogram-based support vector machine: Application in lower-limb active exoskeleton. aug: au: Mazumder, Oishee Kundu, Ananda Sankar Lenka, Prasanna Kumar Bhaumik, Subhasis affil: School of Mechatronics and Robotics, Indian Institute of Engineering Science and Technology, Shibpur, P.O. Botanic Garden, Shalimar, Howrah, West Bengal 711103, India sug: subj: Algorithms Walking Physiology Lower Extremity Physiology Accelerometry Adult Young Adult Pressure Human Adult: 19-44 years ab: Ambulatory activity classification is an active area of research for controlling and monitoring state initiation, termination, and transition in mobility assistive devices such as lower-limb exoskeletons. State transition of lower-limb exoskeletons reported thus far are achieved mostly through the use of manual switches or state machine-based logic. In this paper, we propose a postural activity classifier using a 'dendogram-based support vector machine' (DSVM) which can be used to control a lower-limb exoskeleton. A pressure sensor-based wearable insole and two six-axis inertial measurement units (IMU) have been used for recognising two static and seven dynamic postural activities: sit, stand, and sit-to-stand, stand-to-sit, level walk, fast walk, slope walk, stair ascent and stair descent. Most of the ambulatory activities are periodic in nature and have unique patterns of response. The proposed classification algorithm involves the recognition of activity patterns on the basis of the periodic shape of trajectories. Polynomial coefficients extracted from the hip angle trajectory and the centre-of-pressure (CoP) trajectory during an activity cycle are used as features to classify dynamic activities. The novelty of this paper lies in finding suitable instrumentation, developing post-processing techniques, and selecting shape-based features for ambulatory activity classification. The proposed activity classifier is used to identify the activity states of a lower-limb exoskeleton. The DSVM classifier algorithm achieved an overall classification accuracy of 95.2%. pubtype: Academic Journal doctype: research Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|