Multiclass real-time intent recognition of a powered lower limb prosthesis

Huseyin Atakan Varol, Frank Sup, Michael Goldfarb

Research output: Contribution to journalArticle

186 Citations (Scopus)

Abstract

This paper describes a control architecture and intent recognition approach for the real-time supervisory control of a powered lower limb prosthesis. The approach infers user intent to stand, sit, or walk, by recognizing patterns in prosthesis sensor data in real time, without the need for instrumentation of the sound-side leg. Specifically, the intent recognizer utilizes time-based features extracted from frames of prosthesis signals, which are subsequently reduced to a lower dimensionality (for computational efficiency). These data are initially used to train intent models, which classify the patterns as standing, sitting, or walking. The trained models are subsequently used to infer the user's intent in real time. In addition to describing the generalized control approach, this paper describes the implementation of this approach on a single unilateral transfemoral amputee subject and demonstrates via experiments the effectiveness of the approach. In the real-time supervisory control experiments, the intent recognizer identified all 90 activity-mode transitions, switching the underlying middle-level controllers without any perceivable delay by the user. The intent recognizer also identified six activity-mode transitions, which were not intended by the user. Due to the intentional overlapping functionality of the middle-level controllers, the incorrect classifications neither caused problems in functionality, nor were perceived by the user.

Original languageEnglish
Article number5290091
Pages (from-to)542-551
Number of pages10
JournalIEEE Transactions on Biomedical Engineering
Volume57
Issue number3
DOIs
Publication statusPublished - Mar 2010
Externally publishedYes

Fingerprint

Controllers
Computational efficiency
Experiments
Acoustic waves
Prostheses and Implants
Sensors

Keywords

  • Pattern recognition
  • Physical human-robot interaction
  • Powered prosthesis
  • Rehabilitation robotics

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Multiclass real-time intent recognition of a powered lower limb prosthesis. / Varol, Huseyin Atakan; Sup, Frank; Goldfarb, Michael.

In: IEEE Transactions on Biomedical Engineering, Vol. 57, No. 3, 5290091, 03.2010, p. 542-551.

Research output: Contribution to journalArticle

@article{458ed6dd669d4db0ba68abb35478e00c,
title = "Multiclass real-time intent recognition of a powered lower limb prosthesis",
abstract = "This paper describes a control architecture and intent recognition approach for the real-time supervisory control of a powered lower limb prosthesis. The approach infers user intent to stand, sit, or walk, by recognizing patterns in prosthesis sensor data in real time, without the need for instrumentation of the sound-side leg. Specifically, the intent recognizer utilizes time-based features extracted from frames of prosthesis signals, which are subsequently reduced to a lower dimensionality (for computational efficiency). These data are initially used to train intent models, which classify the patterns as standing, sitting, or walking. The trained models are subsequently used to infer the user's intent in real time. In addition to describing the generalized control approach, this paper describes the implementation of this approach on a single unilateral transfemoral amputee subject and demonstrates via experiments the effectiveness of the approach. In the real-time supervisory control experiments, the intent recognizer identified all 90 activity-mode transitions, switching the underlying middle-level controllers without any perceivable delay by the user. The intent recognizer also identified six activity-mode transitions, which were not intended by the user. Due to the intentional overlapping functionality of the middle-level controllers, the incorrect classifications neither caused problems in functionality, nor were perceived by the user.",
keywords = "Pattern recognition, Physical human-robot interaction, Powered prosthesis, Rehabilitation robotics",
author = "Varol, {Huseyin Atakan} and Frank Sup and Michael Goldfarb",
year = "2010",
month = "3",
doi = "10.1109/TBME.2009.2034734",
language = "English",
volume = "57",
pages = "542--551",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "3",

}

TY - JOUR

T1 - Multiclass real-time intent recognition of a powered lower limb prosthesis

AU - Varol, Huseyin Atakan

AU - Sup, Frank

AU - Goldfarb, Michael

PY - 2010/3

Y1 - 2010/3

N2 - This paper describes a control architecture and intent recognition approach for the real-time supervisory control of a powered lower limb prosthesis. The approach infers user intent to stand, sit, or walk, by recognizing patterns in prosthesis sensor data in real time, without the need for instrumentation of the sound-side leg. Specifically, the intent recognizer utilizes time-based features extracted from frames of prosthesis signals, which are subsequently reduced to a lower dimensionality (for computational efficiency). These data are initially used to train intent models, which classify the patterns as standing, sitting, or walking. The trained models are subsequently used to infer the user's intent in real time. In addition to describing the generalized control approach, this paper describes the implementation of this approach on a single unilateral transfemoral amputee subject and demonstrates via experiments the effectiveness of the approach. In the real-time supervisory control experiments, the intent recognizer identified all 90 activity-mode transitions, switching the underlying middle-level controllers without any perceivable delay by the user. The intent recognizer also identified six activity-mode transitions, which were not intended by the user. Due to the intentional overlapping functionality of the middle-level controllers, the incorrect classifications neither caused problems in functionality, nor were perceived by the user.

AB - This paper describes a control architecture and intent recognition approach for the real-time supervisory control of a powered lower limb prosthesis. The approach infers user intent to stand, sit, or walk, by recognizing patterns in prosthesis sensor data in real time, without the need for instrumentation of the sound-side leg. Specifically, the intent recognizer utilizes time-based features extracted from frames of prosthesis signals, which are subsequently reduced to a lower dimensionality (for computational efficiency). These data are initially used to train intent models, which classify the patterns as standing, sitting, or walking. The trained models are subsequently used to infer the user's intent in real time. In addition to describing the generalized control approach, this paper describes the implementation of this approach on a single unilateral transfemoral amputee subject and demonstrates via experiments the effectiveness of the approach. In the real-time supervisory control experiments, the intent recognizer identified all 90 activity-mode transitions, switching the underlying middle-level controllers without any perceivable delay by the user. The intent recognizer also identified six activity-mode transitions, which were not intended by the user. Due to the intentional overlapping functionality of the middle-level controllers, the incorrect classifications neither caused problems in functionality, nor were perceived by the user.

KW - Pattern recognition

KW - Physical human-robot interaction

KW - Powered prosthesis

KW - Rehabilitation robotics

UR - http://www.scopus.com/inward/record.url?scp=77649160735&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77649160735&partnerID=8YFLogxK

U2 - 10.1109/TBME.2009.2034734

DO - 10.1109/TBME.2009.2034734

M3 - Article

VL - 57

SP - 542

EP - 551

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 3

M1 - 5290091

ER -