Open AccessDetecting sleep in drivers during highly automated driving: the potential of physiological parameters
Author(s) -
Wörle Johanna,
Metz Barbara,
Thiele Christian,
Weller Gert
Publication year - 2019
Publication title -
iet intelligent transport systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.579
H-Index - 45
eISSN - 1751-9578
pISSN - 1751-956X
DOI - 10.1049/iet-its.2018.5529
Subject(s) - wakefulness , electroencephalography , driving simulator , falling (accident) , sleep (system call) , automation , simulation , computer science , fidelity , poison control , set (abstract data type) , physical medicine and rehabilitation , psychology , engineering , neuroscience , medicine , medical emergency , telecommunications , operating system , mechanical engineering , programming language , psychiatry
Driver monitoring is added as a primary safety measure in the EuroNCAP roadmap 2025. Especially with the introduction of automated driving systems into the market, new requirements are set to driver monitoring systems (DMSs). When not being actively involved in driving, the risk of drivers becoming drowsy and even falling asleep at the wheel increases. Modern DMSs will have to be able to detect a driver falling asleep or sleeping in order for the automation to take appropriate actions. Conventional measures for detecting the driver state such as analysing the driving behaviour are not available in automated driving. The aim of the study was to identify potential physiological measures as a basis for the development of systems that are able to detect sleep in drivers during automated driving. A within‐subjects study with N = 21 subjects was conducted in a high‐fidelity driving simulator. Electromyography, electrodermal activity (EDA), respiration and electrocardiography (ECG) were measured in drivers during states of wakefulness and sleep. Sleep stages were assigned with the electroencephalography as a ground truth. The results indicate the potential of EDA and ECG parameters to differentiate between sleep and wakefulness. Implications for the implementation in DMS are discussed.