Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers

Thomas Cano Hald, David Holmgaard Junker, Mads Mårtensson, Mikael Skov, Dimitrios Raptis

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

1 Citation (Scopus)

Resumé

People increasingly interact with social media or other apps on their smartphones while driving car. This is naturally a major safety concern, and it remains unclear how to avoid or limit such interaction. We investigate this problem through human activity recognition (HAR) where we developed a system called IRIS, which collects smartwatch accelerometer data and analyses the data through machine learning and predicts if the data origins from a driver or a passenger. We report from a field experiment with 24 participants acting as drivers or passengers where we achieved an overall prediction accuracy of 87%. We further found that various road segments had less effect on the accuracy than anticipated, but we also found that passenger tasks had a negative effect on recognition accuracy. We discuss several implications from findings.
OriginalsprogEngelsk
TitelProceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications
Antal sider11
ForlagAssociation for Computing Machinery
Publikationsdato23 sep. 2018
Sider74-84
ISBN (Elektronisk)978-1-4503-5946-7
DOI
StatusUdgivet - 23 sep. 2018
Begivenhed10th International ACM Conference on Automotive User Interfaces - Toronto, Canada
Varighed: 23 sep. 201825 sep. 2018

Konference

Konference10th International ACM Conference on Automotive User Interfaces
LandCanada
ByToronto
Periode23/09/201825/09/2018

Fingerprint

Smartphones
Accelerometers
Application programs
Learning systems
Railroad cars
Sensors
Experiments

Citer dette

Cano Hald, T., Junker, D. H., Mårtensson, M., Skov, M., & Raptis, D. (2018). Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers. I Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (s. 74-84). Association for Computing Machinery. https://doi.org/10.1145/3239060.3239068
Cano Hald, Thomas ; Junker, David Holmgaard ; Mårtensson, Mads ; Skov, Mikael ; Raptis, Dimitrios. / Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers. Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications. Association for Computing Machinery, 2018. s. 74-84
@inproceedings{a718c6021d1148c0a47e5dc46f1c9ccd,
title = "Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers",
abstract = "People increasingly interact with social media or other apps on their smartphones while driving car. This is naturally a major safety concern, and it remains unclear how to avoid or limit such interaction. We investigate this problem through human activity recognition (HAR) where we developed a system called IRIS, which collects smartwatch accelerometer data and analyses the data through machine learning and predicts if the data origins from a driver or a passenger. We report from a field experiment with 24 participants acting as drivers or passengers where we achieved an overall prediction accuracy of 87{\%}. We further found that various road segments had less effect on the accuracy than anticipated, but we also found that passenger tasks had a negative effect on recognition accuracy. We discuss several implications from findings.",
keywords = "Accelerometer data, Driving, Human activity recognition, Sensor data, Smartwatch",
author = "{Cano Hald}, Thomas and Junker, {David Holmgaard} and Mads M{\aa}rtensson and Mikael Skov and Dimitrios Raptis",
year = "2018",
month = "9",
day = "23",
doi = "10.1145/3239060.3239068",
language = "English",
pages = "74--84",
booktitle = "Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications",
publisher = "Association for Computing Machinery",
address = "United States",

}

Cano Hald, T, Junker, DH, Mårtensson, M, Skov, M & Raptis, D 2018, Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers. i Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications. Association for Computing Machinery, s. 74-84, Toronto, Canada, 23/09/2018. https://doi.org/10.1145/3239060.3239068

Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers. / Cano Hald, Thomas; Junker, David Holmgaard; Mårtensson, Mads; Skov, Mikael; Raptis, Dimitrios.

Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications. Association for Computing Machinery, 2018. s. 74-84.

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers

AU - Cano Hald, Thomas

AU - Junker, David Holmgaard

AU - Mårtensson, Mads

AU - Skov, Mikael

AU - Raptis, Dimitrios

PY - 2018/9/23

Y1 - 2018/9/23

N2 - People increasingly interact with social media or other apps on their smartphones while driving car. This is naturally a major safety concern, and it remains unclear how to avoid or limit such interaction. We investigate this problem through human activity recognition (HAR) where we developed a system called IRIS, which collects smartwatch accelerometer data and analyses the data through machine learning and predicts if the data origins from a driver or a passenger. We report from a field experiment with 24 participants acting as drivers or passengers where we achieved an overall prediction accuracy of 87%. We further found that various road segments had less effect on the accuracy than anticipated, but we also found that passenger tasks had a negative effect on recognition accuracy. We discuss several implications from findings.

AB - People increasingly interact with social media or other apps on their smartphones while driving car. This is naturally a major safety concern, and it remains unclear how to avoid or limit such interaction. We investigate this problem through human activity recognition (HAR) where we developed a system called IRIS, which collects smartwatch accelerometer data and analyses the data through machine learning and predicts if the data origins from a driver or a passenger. We report from a field experiment with 24 participants acting as drivers or passengers where we achieved an overall prediction accuracy of 87%. We further found that various road segments had less effect on the accuracy than anticipated, but we also found that passenger tasks had a negative effect on recognition accuracy. We discuss several implications from findings.

KW - Accelerometer data

KW - Driving

KW - Human activity recognition

KW - Sensor data

KW - Smartwatch

UR - http://www.scopus.com/inward/record.url?scp=85055981566&partnerID=8YFLogxK

U2 - 10.1145/3239060.3239068

DO - 10.1145/3239060.3239068

M3 - Article in proceeding

SP - 74

EP - 84

BT - Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications

PB - Association for Computing Machinery

ER -

Cano Hald T, Junker DH, Mårtensson M, Skov M, Raptis D. Using Smartwatch Inertial Sensors to Recognize and Distinguish Between Car Drivers and Passengers. I Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications. Association for Computing Machinery. 2018. s. 74-84 https://doi.org/10.1145/3239060.3239068