An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP

Søren Møller Christensen, Nicklas Stubkjær Holm, Sadasivan Puthusserypady

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

165 Downloads (Pure)

Abstract

Worldwide, millions of people are locked in or in a wheelchair, due to several neuromuscular disorders or spinal cord injuries. These individuals are deprived of trivial social activities, like interacting or playing games with other people. Such activities are crucial for personal development, and can have a great impact on the quality of their lives. This work aims at the design and implementation of an electroencephalography (EEG) based motor imagery (MI) brain computer interface (BCI) system that would allow disabled, and able-bodied, individuals alike to control a drone in a 3D physical environment by only using their thoughts. An improved version of the filter bank common spatial pattern (FBCSP) algorithm was developed, and it has shown to perform superior (68.5% accuracy) to the winning FBCSP algorithm (67.8% accuracy), when tested on dataset 2a (4 class MI) of the BCI competition IV. A deep convolutional neural network (CNN) based algorithm was also implemented and tested on the same dataset, which however performed inferior (62.9% accuracy) to the winner, as well as our proposed FBCSP algorithms. The improved FBCSP was then tested on our inhouse 5-class (left hand, right hand, tongue, both feet and rest) MI dataset (collected from 10 able-bodied subjects) and obtained a mean accuracy of 41.8±11.74%. This is considered a significant result though it is not good enough to attempt the control of a real drone.
Original languageEnglish
Title of host publicationProceedings of 2019 7th International Winter Conference on Brain-Computer Interface
Number of pages6
PublisherIEEE
Publication date2019
ISBN (Print)978-1-5386-8116-9
DOIs
Publication statusPublished - 2019
Event7th International Winter Conference on Brain-Computer Interface - High 1 Resort, Jeongseon, Korea, Republic of
Duration: 18 Feb 201920 Feb 2019
http://brain.korea.ac.kr/bci2019/

Conference

Conference7th International Winter Conference on Brain-Computer Interface
LocationHigh 1 Resort
CountryKorea, Republic of
CityJeongseon
Period18/02/201920/02/2019
Internet address

Keywords

  • Brain Computer Interface (BCI)
  • Motor Imagery (MI)
  • Filter Bank Common Spatial Pattern (FBCSP)
  • Convolutional Neural Network (CNN)
  • Drone Control
  • Multiclass Motor Imagery

Cite this

Christensen, S. M., Holm, N. S., & Puthusserypady, S. (2019). An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP. In Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface IEEE. https://doi.org/10.1109/IWW-BCI.2019.8737263
Christensen, Søren Møller ; Holm, Nicklas Stubkjær ; Puthusserypady, Sadasivan. / An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP. Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface. IEEE, 2019.
@inproceedings{087c0c53815a4a31a27c399688f6fc9d,
title = "An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP",
abstract = "Worldwide, millions of people are locked in or in a wheelchair, due to several neuromuscular disorders or spinal cord injuries. These individuals are deprived of trivial social activities, like interacting or playing games with other people. Such activities are crucial for personal development, and can have a great impact on the quality of their lives. This work aims at the design and implementation of an electroencephalography (EEG) based motor imagery (MI) brain computer interface (BCI) system that would allow disabled, and able-bodied, individuals alike to control a drone in a 3D physical environment by only using their thoughts. An improved version of the filter bank common spatial pattern (FBCSP) algorithm was developed, and it has shown to perform superior (68.5{\%} accuracy) to the winning FBCSP algorithm (67.8{\%} accuracy), when tested on dataset 2a (4 class MI) of the BCI competition IV. A deep convolutional neural network (CNN) based algorithm was also implemented and tested on the same dataset, which however performed inferior (62.9{\%} accuracy) to the winner, as well as our proposed FBCSP algorithms. The improved FBCSP was then tested on our inhouse 5-class (left hand, right hand, tongue, both feet and rest) MI dataset (collected from 10 able-bodied subjects) and obtained a mean accuracy of 41.8±11.74{\%}. This is considered a significant result though it is not good enough to attempt the control of a real drone.",
keywords = "Brain Computer Interface (BCI), Motor Imagery (MI), Filter Bank Common Spatial Pattern (FBCSP), Convolutional Neural Network (CNN), Drone Control, Multiclass Motor Imagery",
author = "Christensen, {S{\o}ren M{\o}ller} and Holm, {Nicklas Stubkj{\ae}r} and Sadasivan Puthusserypady",
year = "2019",
doi = "10.1109/IWW-BCI.2019.8737263",
language = "English",
isbn = "978-1-5386-8116-9",
booktitle = "Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface",
publisher = "IEEE",
address = "United States",

}

Christensen, SM, Holm, NS & Puthusserypady, S 2019, An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP. in Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface. IEEE, 7th International Winter Conference on Brain-Computer Interface, Jeongseon, Korea, Republic of, 18/02/2019. https://doi.org/10.1109/IWW-BCI.2019.8737263

An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP. / Christensen, Søren Møller; Holm, Nicklas Stubkjær; Puthusserypady, Sadasivan.

Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface. IEEE, 2019.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP

AU - Christensen, Søren Møller

AU - Holm, Nicklas Stubkjær

AU - Puthusserypady, Sadasivan

PY - 2019

Y1 - 2019

N2 - Worldwide, millions of people are locked in or in a wheelchair, due to several neuromuscular disorders or spinal cord injuries. These individuals are deprived of trivial social activities, like interacting or playing games with other people. Such activities are crucial for personal development, and can have a great impact on the quality of their lives. This work aims at the design and implementation of an electroencephalography (EEG) based motor imagery (MI) brain computer interface (BCI) system that would allow disabled, and able-bodied, individuals alike to control a drone in a 3D physical environment by only using their thoughts. An improved version of the filter bank common spatial pattern (FBCSP) algorithm was developed, and it has shown to perform superior (68.5% accuracy) to the winning FBCSP algorithm (67.8% accuracy), when tested on dataset 2a (4 class MI) of the BCI competition IV. A deep convolutional neural network (CNN) based algorithm was also implemented and tested on the same dataset, which however performed inferior (62.9% accuracy) to the winner, as well as our proposed FBCSP algorithms. The improved FBCSP was then tested on our inhouse 5-class (left hand, right hand, tongue, both feet and rest) MI dataset (collected from 10 able-bodied subjects) and obtained a mean accuracy of 41.8±11.74%. This is considered a significant result though it is not good enough to attempt the control of a real drone.

AB - Worldwide, millions of people are locked in or in a wheelchair, due to several neuromuscular disorders or spinal cord injuries. These individuals are deprived of trivial social activities, like interacting or playing games with other people. Such activities are crucial for personal development, and can have a great impact on the quality of their lives. This work aims at the design and implementation of an electroencephalography (EEG) based motor imagery (MI) brain computer interface (BCI) system that would allow disabled, and able-bodied, individuals alike to control a drone in a 3D physical environment by only using their thoughts. An improved version of the filter bank common spatial pattern (FBCSP) algorithm was developed, and it has shown to perform superior (68.5% accuracy) to the winning FBCSP algorithm (67.8% accuracy), when tested on dataset 2a (4 class MI) of the BCI competition IV. A deep convolutional neural network (CNN) based algorithm was also implemented and tested on the same dataset, which however performed inferior (62.9% accuracy) to the winner, as well as our proposed FBCSP algorithms. The improved FBCSP was then tested on our inhouse 5-class (left hand, right hand, tongue, both feet and rest) MI dataset (collected from 10 able-bodied subjects) and obtained a mean accuracy of 41.8±11.74%. This is considered a significant result though it is not good enough to attempt the control of a real drone.

KW - Brain Computer Interface (BCI)

KW - Motor Imagery (MI)

KW - Filter Bank Common Spatial Pattern (FBCSP)

KW - Convolutional Neural Network (CNN)

KW - Drone Control

KW - Multiclass Motor Imagery

U2 - 10.1109/IWW-BCI.2019.8737263

DO - 10.1109/IWW-BCI.2019.8737263

M3 - Article in proceedings

SN - 978-1-5386-8116-9

BT - Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface

PB - IEEE

ER -

Christensen SM, Holm NS, Puthusserypady S. An Improved Five Class MI Based BCI Scheme for Drone Control Using Filter Bank CSP. In Proceedings of 2019 7th International Winter Conference on Brain-Computer Interface. IEEE. 2019 https://doi.org/10.1109/IWW-BCI.2019.8737263