Driver fatigue transition prediction in highly automated driving using physiological features

Zhou, Feng; Alsaid, Areen; Blommer, Mike; Curry, Reates; Swaminathan, Radhakrishnan; Kochhar, Dev S.; Talamonti, Walter; Tijerina, Louis; Lei, Baiying

doi:10.1016/j.eswa.2020.113204

Cited by 74 publications

(40 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As indicated by the results of model accuracy, F1-score, and ROC curve comparisons, the random forest approach outperformed the other classification approaches. Consistent with previous studies on drivers' fatigue and drowsiness detection (McDonald et al, 2014;Zhou, Alsaid, et al, 2020), the random forest approach also showed its supremacy for takeover performance prediction. It might be because random forests aggregate the results of many bootstrap aggregated (bagged) decision trees, which reduces the effects of overfitting and improves generalization.…”

Section: Model Performance Comparisonssupporting

confidence: 87%

“…Heart rate (HR) and heart rate variability (HRV) have both been used for assessing drivers' workload in real time (Mehler, Reimer, & Coughlin, 2012;Mehler, Reimer, Coughlin, & Dusek, 2009;Zhou, Alsaid, et al, 2020). Galvanic skin responses (GSRs) were found to reflect drivers' mental activities, and their properties (amplitude, frequency) were used to indicate drivers' changes of arousal related to events (Collet, Clarion, Morel, Chapon, & Petit, 2009).…”

Section: Predicting Drivers' States Through Physiological Measurementsmentioning

confidence: 99%

“…Third, our study employed a random forest model in addition to the machine learning models used in previous studies to predict takeover performance. Random forests have been proved to have great prediction performance for classification problems (Dietterich, 1997;McDonald, Lee, Schwarz, & Brown, 2014;Zhou, Alsaid, et al, 2020).…”

Section: The Present Studymentioning

confidence: 99%

See 2 more Smart Citations

Predicting driver takeover performance in conditionally automated driving

Zhou

Pulver

et al. 2020

Accident Analysis & Prevention

Self Cite

View full text Add to dashboard Cite

In conditionally automated driving, drivers have difficulty taking over control when requested. To address this challenge, we aimed to predict drivers' takeover performance before the issue of a takeover request (TOR) by analyzing drivers' physiological data and external environment data. We used data sets from two human-in-the-loop experiments, wherein drivers engaged in non-driving-related tasks (NDRTs) were requested to take over control from automated driving in various situations. Drivers' physiological data included heart rate indices, galvanic skin response indices, and eye-tracking metrics. Driving environment data included scenario type, traffic density, and TOR lead time. Drivers' takeover performance was categorized as good or bad according to their driving behaviors during the transition period and was treated as the ground truth. Using six machine learning methods, we found that the random forest classifier performed the best and was able to predict drivers' takeover performance when they were engaged in NDRTs with different levels of cognitive load. We recommended 3 s as the optimal time window to predict takeover performance using the random forest classifier, with an accuracy of 84.3% and an F1-score of 64.0%. Our findings have implications for the algorithm development of driver state detection and the design of adaptive in-vehicle alert systems in conditionally automated driving.

show abstract

Section: Model Performance Comparisonssupporting

confidence: 87%

Section: Predicting Drivers' States Through Physiological Measurementsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting driver takeover performance in conditionally automated driving

Zhou

Pulver

et al. 2020

Accident Analysis & Prevention

Self Cite

View full text Add to dashboard Cite

show abstract

“…Future research should examine whether using multitasking measures provide more reliable assessments of cognitive fatigue as compared to single-task measures. It is also important to note that these measures are only able to detect cognitive fatigue after a considerable decline in behavioural performance, which can often be detected too late in critical situations [ 37 ].…”

Section: Traditional Psychological Assessments Of Cognitive Fatiguementioning

confidence: 99%

Biomarker-Informed Machine Learning Model of Cognitive Fatigue from a Heart Rate Response Perspective

Lee

Gan

Christopoulos

2021

Sensors

View full text Add to dashboard Cite

Cognitive fatigue is a psychological state characterised by feelings of tiredness and impaired cognitive functioning arising from high cognitive demands. This paper examines the recent research progress on the assessment of cognitive fatigue and provides informed recommendations for future research. Traditionally, cognitive fatigue is introspectively assessed through self-report or objectively inferred from a decline in behavioural performance. However, more recently, researchers have attempted to explore the biological underpinnings of cognitive fatigue to understand and measure this phenomenon. In particular, there is evidence indicating that the imbalance between sympathetic and parasympathetic nervous activity appears to be a physiological correlate of cognitive fatigue. This imbalance has been indexed through various heart rate variability indices that have also been proposed as putative biomarkers of cognitive fatigue. Moreover, in contrast to traditional inferential methods, there is also a growing research interest in using data-driven approaches to assessing cognitive fatigue. The ubiquity of wearables with the capability to collect large amounts of physiological data appears to be a major facilitator in the growth of data-driven research in this area. Preliminary findings indicate that such large datasets can be used to accurately predict cognitive fatigue through various machine learning approaches. Overall, the potential of combining domain-specific knowledge gained from biomarker research with machine learning approaches should be further explored to build more robust predictive models of cognitive fatigue.

show abstract

“…However, too many subjective factors probably lead to inaccurate results. The objective evaluation methods are divided into three categories, namely, detection based on driver physiological parameters [4][5][6][7][8], detection based on vehicle behavior [9,10], and detection based on computer vision [11][12][13][14][15][16][17][18][19][20]. The detection based on driver physiological parameters, including electroencephalogram (EEG), electrocardiogram (ECG), electromyography (EMG), electrooculogram (EOG), and other parameters, can reflect the driver's physiological state.…”

Section: Introductionmentioning

confidence: 99%

Driver Fatigue Tracking and Detection Method Based on OpenMV

Zhou¹,

Gong²,

Qie³

et al. 2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Aiming at the problem of fatigue driving, this paper proposed a driver fatigue tracking and detection method combined with OpenMV. OpenMV is used for image acquisition, and the Dlib feature point model is used to locate the detected driver's face. The aspect ratio of eyes is calculated to judge the opening and closing of eyes, and then fatigue detection is performed by PERCLOS (Percentage of Eyelid Closure over the Pupil). The cradle head system is mounted for driver fatigue tracking and detection. The results show that the dynamic tracking system can improve the accuracy to 92.10% for relaxation and 85.20% for fatigue of driver fatigue detection. This system can be applied to the monitoring of fatigue driving very effectively.

show abstract

Driver fatigue transition prediction in highly automated driving using physiological features

Cited by 74 publications

References 26 publications

Predicting driver takeover performance in conditionally automated driving

Predicting driver takeover performance in conditionally automated driving

Biomarker-Informed Machine Learning Model of Cognitive Fatigue from a Heart Rate Response Perspective

Driver Fatigue Tracking and Detection Method Based on OpenMV

Contact Info

Product

Resources

About