LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor

Christopher Millar; N Siddique; Emmett Kerr

doi:10.1109/ICCAR52225.2021.9463477

LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor

Christopher Millar
, N Siddique
, Emmett Kerr

Research output: Contribution to conference › Paper › peer-review

5 Citations (Scopus)

154 Downloads (Pure)

Abstract

Modelling human grasping and transferring this data to an anthropomorphic robotic hand to endow it with human like grasping capabilities is a complex task. In this paper the use of surface electromyography (sEMG) for classification of functional grasps associated with everyday life is carried out using a low-cost wearable sensor in conjunction with state-of-the-art recurrent neural networks. The results produced through these experiments demonstrate the potential for sEMG to be used as an effective medium for transferring human demonstration to a robotic system.

Original language	English
Pages	213-222
Number of pages	10
DOIs	https://doi.org/10.1109/ICCAR52225.2021.9463477
Publication status	Published (in print/issue) - 25 Jun 2021
Event	7th Annual International Conference on Control, Automation and Robotics - Singapore Duration: 23 Apr 2021 → 26 Apr 2021 http://iccar.org/index.html

Conference

Conference	7th Annual International Conference on Control, Automation and Robotics
Abbreviated title	ICCAR
Period	23/04/21 → 26/04/21
Internet address	http://iccar.org/index.html

Bibliographical note

Publisher Copyright:
© 2021 IEEE.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Keywords

Recurrent Neural Network
Pattern Classification
Grasping
Muscles
Robot Sensing Systems
Task Analysis
Training
grasping
LSTM
signal classification
recurrent neural networks
sEMG
wearable sensors

Access to Document

10.1109/ICCAR52225.2021.9463477

Chris Millar accepted vresionAuthor Accepted Manuscript, 8.26 MB

Cite this

@conference{358acfcf791740b7bce037882fee0d6e,

title = "LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor",

abstract = "Modelling human grasping and transferring this data to an anthropomorphic robotic hand to endow it with human like grasping capabilities is a complex task. In this paper the use of surface electromyography (sEMG) for classification of functional grasps associated with everyday life is carried out using a low-cost wearable sensor in conjunction with state-of-the-art recurrent neural networks. The results produced through these experiments demonstrate the potential for sEMG to be used as an effective medium for transferring human demonstration to a robotic system.",

keywords = "Recurrent Neural Network, Pattern Classification, Grasping, Muscles, Robot Sensing Systems, Task Analysis, Training, grasping, LSTM, signal classification, recurrent neural networks, sEMG, wearable sensors",

author = "Christopher Millar and N Siddique and Emmett Kerr",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 7th Annual International Conference on Control, Automation and Robotics, ICCAR ; Conference date: 23-04-2021 Through 26-04-2021",

year = "2021",

month = jun,

day = "25",

doi = "10.1109/ICCAR52225.2021.9463477",

language = "English",

pages = "213--222",

url = "http://iccar.org/index.html",

}

TY - CONF

T1 - LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor

AU - Millar, Christopher

AU - Siddique, N

AU - Kerr, Emmett

PY - 2021/6/25

Y1 - 2021/6/25

N2 - Modelling human grasping and transferring this data to an anthropomorphic robotic hand to endow it with human like grasping capabilities is a complex task. In this paper the use of surface electromyography (sEMG) for classification of functional grasps associated with everyday life is carried out using a low-cost wearable sensor in conjunction with state-of-the-art recurrent neural networks. The results produced through these experiments demonstrate the potential for sEMG to be used as an effective medium for transferring human demonstration to a robotic system.

AB - Modelling human grasping and transferring this data to an anthropomorphic robotic hand to endow it with human like grasping capabilities is a complex task. In this paper the use of surface electromyography (sEMG) for classification of functional grasps associated with everyday life is carried out using a low-cost wearable sensor in conjunction with state-of-the-art recurrent neural networks. The results produced through these experiments demonstrate the potential for sEMG to be used as an effective medium for transferring human demonstration to a robotic system.

KW - Recurrent Neural Network

KW - Pattern Classification

KW - Grasping

KW - Muscles

KW - Robot Sensing Systems

KW - Task Analysis

KW - Training

KW - grasping

KW - LSTM

KW - signal classification

KW - recurrent neural networks

KW - sEMG

KW - wearable sensors

UR - https://www.scopus.com/pages/publications/85114498563

U2 - 10.1109/ICCAR52225.2021.9463477

DO - 10.1109/ICCAR52225.2021.9463477

M3 - Paper

SP - 213

EP - 222

T2 - 7th Annual International Conference on Control, Automation and Robotics

Y2 - 23 April 2021 through 26 April 2021

ER -

LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor

Abstract

Conference

Bibliographical note

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

LSTM Network Classification of Dexterous Individual Finger Movements

Kinaesthetic learning through sEMG signal classification

Cite this

LSTM Classification of Functional Grasps Using sEMG Data from Low-Cost Wearable Sensor

Abstract

Conference

Bibliographical note

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Research output

LSTM Network Classification of Dexterous Individual Finger Movements

Student theses

Kinaesthetic learning through sEMG signal classification

Cite this