The objective of this paper is to investigate if signal analysis and machine learning can be used to develop an accurate sleepiness warning system. The developed system was trained using the supposedly most reliable sleepiness indicators available, extracted from electroencephalography, electrocardiography, electrooculography, and driving performance data (steering behavior and lane positioning). Sequential forward floating selection was used to select the most descriptive features, and five different classifiers were tested. A unique data set with 86 drivers, obtained while driving on real roads in real traffic, both in alert and sleep deprived conditions, was used to train and test the classifiers. Subjective ratings using the Karolinska sleepiness scale (KSS) was used to split the data as sufficiently alert (KSS ≤ 6) or sleepy (KSS ≥ 8). KSS = 7 was excluded to get a clearer distinction between the groups. A random forest classifier was found to be the most robust classifier with an accuracy of 94.1% (sensitivity 86.5%, specificity 95.7%). The results further showed the importance of personalizing a sleepiness detection system. When testing the classifier on data from a person that it had not been trained on, the sensitivity dropped to 41.4%. One way to improve the sensitivity was to add a biomathematical model of sleepiness amongst the features, which increased the sensitivity to 66.2% for participant-independent classification. Future works include taking contextual features into account, using classifiers that takes full advantage of sequential data, and to develop models that adapt to individual drivers.

Driver Sleepiness Classification Based on Physiological Data and Driving Performance From Real Road Driving

Obstructive sleep apnoea (OSA) is highly prevalent and is a recognised risk factor for motor vehicle accidents (MVA). Effective treatment with continuous positive airway pressure (CPAP) has been associated with a normalisation of this increased accident risk. Thus, many jurisdictions have introduced regulations restricting the ability of OSA patients from driving until effectively treated. However, uncertainty prevails regarding the relative importance of OSA severity determined by the apnoea-hypopnoea frequency per hour and the degree of sleepiness in determining accident risk. Furthermore, the identification of subjects at risk for OSA and/or accident risk remains elusive. The introduction of official European regulations regarding fitness to drive prompted the European Respiratory Society to establish a Task Force to address the topic of sleep apnoea, sleepiness and driving with a view to providing an overview to clinicians involved in treating patients with the disorder. The present report evaluates the epidemiology of MVA in patients with OSA, the mechanisms involved in this association, the role of screening questionnaires, driving simulators and other techniques to evaluate sleepiness and/or impaired vigilance, the impact of treatment on MVA risk in affected drivers, and highlights the evidence gaps regarding the identification of OSA patients at risk for MVA.

https://erj.ersjournals.com/content/erj/57/2/2001272.full.pdf

European Respiratory Society statement on sleep apnoea, sleepiness and driving risk.

The progression of vehicle active safety research and its implementation have brought a lot of reduction in the number of road fatalities globally. It enables the vehicle to aid the driver in the risky scenarios. One of the examples is Autonomous Emergency Braking System (AEB). It yields the required braking torques intervention in the hazardous events to mitigate collision risks. However, statistics show that a large number of road accidents remain, including the case of near-miss incidents. In this work, to aid the reduction of the near-miss accident, and assist the AEB to maintain the safe-distance with the frontal obstacle, AEB is integrated with the Potential Field (PF) risk assessment strategy. When the PF threshold of the frontal obstacle's risk is violated, AEB provides the active braking intervention. The proposed design was tested and validated using a test vehicle, where it mitigates the collision risk with a static obstacle in a medium-speed scenario. The work is part of the continuous effort to develop the autonomous vehicle and reduce the high number of road accidents in Malaysia. Results show the integration of AEB and PF successfully aid the vehicle to mitigate the collision as well as maintaining the safe distance with the frontal obstacle.

Autonomous emergency braking system with potential field risk assessment for frontal collision mitigation

Driver sleepiness is a contributing factor in many road fatalities. A long-standing goal in driver state research has therefore been to develop a robust sleepiness detection system. It has been suggested that various heart rate variability (HRV) metrics can be used for driver sleepiness classification. However, since heart rate is modulated not only by sleepiness but also by several other time-varying intra-individual factors such as posture, distress, boredom and relaxation, it is relevant to highlight not only the possibilities but also the difficulties involved in HRV-based driver sleepiness classification. This paper investigates the reliability of HRV as a standalone feature for driver sleepiness detection in a realistic setting. Data from three real-road driving studies were used, including 86 drivers in both alert and sleep-deprived conditions. Subjective ratings based on the Karolinska sleepiness scale (KSS) were used as ground truth when training four binary classifiers ( ${k}$ -nearest neighbours, support vector machine, AdaBoost, and random forest). The best performance was achieved with the random forest classifier with an accuracy of 85%. However, the accuracy dropped to 64% for three-class classification and to 44% for subject-independent, leave-one-participant-out classification. The worst results were obtained in the severely sleepy class. The results show that in realistic driving conditions, subject-independent sleepiness classification based on HRV is poor. The conclusion is that more work is needed to control for the many confounding factors that also influence HRV before it can be used as input to a driver sleepiness detection system.

Heart Rate Variability for Classification of Alert Versus Sleep Deprived Drivers in Real Road Driving Conditions

Excessive daytime sleepiness (EDS) while driving is a major international public health issue resulting in a more than doubled risk of motor vehicle accidents (MVAs). Obstructive sleep apnea (OSA) is the most frequent medical cause of EDS. Therefore, the European Union Directive 2014/85/EU determined that “untreated moderate to severe OSA coincident with EDS constitutes a medical disorder leading to unfitness to drive.” The paper aims are to provide a brief review of sleepiness and its implications for driving safety, as well as to describe the subjective and objective methods to accurately evaluate EDS in order to assess fitness to drive in patients with OSA. We examined databases including PubMed, Medline, and EMBASE using the search terms “sleepiness at the wheel, excessive daytime sleepiness, sleepiness measure, sleep-wake cycle, obstructive sleep apnea, driving license, fitness to drive.” Significant interindividual variability in EDS exists in patients with comparable severity of OSA. Objective methods of measuring EDS are too expensive and time consuming to be suitable for the certification of driving licenses. The reliability of subjective methods depends upon the clinical setting and subjective tools assess only limited aspects of EDS. Objective measures, such as biochemical biomarkers, must, therefore, support subjective methods. Extensive data have supported different subjective and objective methods for the appraisal of EDS in patients with OSA depending upon the clinical and experimental setting. Challenges remain to determine an appropriate tool for the evaluation of fitness to drive.

Daniel Zwahlen

Papers

Sleepiness, driving, and motor vehicle accidents: A questionnaire-based survey

Tödliche Luftembolie durch einvernehmlichen Geschlechtsverkehr

Fatal air embolism due to consensual sexual intercourse