Multimodal driver state modeling through unsupervised learning

Tavakoli, Arash; Heydarian, Arsalan

doi:10.1016/j.aap.2022.106640

Cited by 20 publications

(8 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the sudden change in the environment at the entrance of the tunnel, it is very easy to cause discomfort and abnormal driving behavior in the physiological and psychological aspects of the driver. Tavakoli A et al [1] detected four patterns of driving behavior (i.e., hard braking, normal braking, corner driving, and highway driving) and two patterns of driver's heart rate (i.e., normal versus abnormally high heart rate) in two case studies. The results of these case studies indicated that a higher percentage of participants had abnormal patterns of driver HR during hard braking, acceleration, and curve driving.…”

Section: Introductionmentioning

confidence: 99%

HR-based analysis of driving behavior characteristics at the entrance section of high-altitude road tunnels

Zhu

Wang²,

Zhang³

2023

Sixth International Conference on Traffic Engineering and Transportation System (ICTETS 2022)

View full text Add to dashboard Cite

To study the driving behaviors of drivers at the entrance section of high-altitude highway tunnels, the data related to drivers' heart rate and longitudinal acceleration of vehicles were obtained through real vehicle tests. To study the driving behavior at the entrance section of high-altitude highway tunnels, the data related to the driver's heart rate and vehicle longitudinal acceleration were obtained through real vehicle tests. Then, the acceleration standard deviation algorithm was used to calculate the standard deviation values of longitudinal acceleration in the entrance section of the high-altitude tunnel, the normal tunnel section, and the highway section, and to compare and analyze them. The results show that when driving in a high-altitude highway tunnel driving environment, in the tunnel entrance section, the longitudinal acceleration changes significantly with the HR, which shows a deceleration behavior when approaching the tunnel entrance and speeding up behavior after entering the tunnel; secondly, from the longitudinal acceleration standard deviation comparison, it can be seen that the longitudinal acceleration standard deviation is the largest in the high-altitude tunnel entrance section, then in the ordinary tunnel section, while the highway section longitudinal The acceleration standard deviation is the smallest, which indicates that the driver's operation of the vehicle is higher than other road sections when passing through the high-altitude entrance section.

show abstract

Section: Introductionmentioning

confidence: 99%

HR-based analysis of driving behavior characteristics at the entrance section of high-altitude road tunnels

Zhu

Wang²,

Zhang³

2023

Sixth International Conference on Traffic Engineering and Transportation System (ICTETS 2022)

View full text Add to dashboard Cite

show abstract

“…In their study, the authors found out that being followed by a vehicle, following a lead vehicle too closely, arriving at an intersection, and performing secondary tasks might be associated with abrupt increases in drivers' HR, possibly showing stress and negative emotions. Another study performed by [31] analyzed drivers' psychophysiological measures (HR and gaze entropy) through a naturalistic study. In their study authors found out that drivers' had higher fraction of normal HR patterns as well as lower gaze entropy pattern within highway driving.…”

Section: Introductionmentioning

confidence: 99%

How are Drivers' Stress Levels and Emotions Associated with the Driving Context? A Naturalistic Study

Tavakoli¹,

Lai²,

Balali³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Understanding and mitigating drivers' negative emotions, stress levels, and anxiety is of high importance for decreasing accident rates, enhancing road safety, and providing a healthy lifestyle to the community of drivers. While detecting drivers' stress and negative emotions can significantly help with this goal, understanding what might be associated with increases in drivers' negative emotions and high stress level, might better help with planning interventions. While studies have provided significant insight into detecting drivers' emotions and stress levels; not many studies focused on the reasons behind changes in stress levels and negative emotions. In this study, by using a naturalistic driving study database, we analyze the changes in the driving scene, including road objects and the dynamical relationship between the ego vehicle and the lead vehicle with respect to changes in drivers' psychophysiological metrics (i.e., heart rate (HR) and facial expressions). We find that different road objects might be associated with varying levels of increase in drivers' HR as well as different proportions of negative facial emotions detected through computer vision. Our results indicate that larger vehicles on the road, such as trucks and buses, are associated with the highest amount of increase in drivers' HR as well as negative emotions. Additionally, we provide evidence that shorter distances to the lead vehicle in naturalistic driving, as well as the higher standard deviation in the distance, might be associated with a higher number of abrupt increases in drivers' HR, depicting a possible increase in stress level. Lastly, our results indicate more positive emotions, lower facial engagement, and a lower abrupt increase in HR at a higher speed of driving, which often happens in highway driving. Our results collectively show that driving at higher speeds happening in highways by avoiding certain road objects might be a better fit for keeping drivers in a calmer, more positive state. This research lays the ground for designing human-centered vehicles, urban environments, and services that can understand the level to which each road object and environment might affect drivers' and passengers' wellbeing.

show abstract

“…As road traffic accidents are among the leading contributors of injuries and fatalities, the J o u r n a l P r e -p r o o f development of traffic accident analysis and prediction models is an important field of research (Chand, Jayesh, & Bhasi, 2021;Sangkharat et al, 2021;Tavakoli & Heydarian, 2022). All traffic accidents have a cost, and this cost can even be immeasurable when human injuries and fatalities are involved (French & Gumus, 2009;Kaygisiz, Senbil, & Yildiz, 2017).…”

Section: Introductionmentioning

confidence: 99%

A knowledge elicitation approach to traffic accident analysis in open data: comparing periods before and after the Covid-19 outbreak

Kao

Chang

2022

Heliyon

View full text Add to dashboard Cite

Multimodal driver state modeling through unsupervised learning

Cited by 20 publications

References 56 publications

HR-based analysis of driving behavior characteristics at the entrance section of high-altitude road tunnels

HR-based analysis of driving behavior characteristics at the entrance section of high-altitude road tunnels

How are Drivers' Stress Levels and Emotions Associated with the Driving Context? A Naturalistic Study

A knowledge elicitation approach to traffic accident analysis in open data: comparing periods before and after the Covid-19 outbreak

Contact Info

Product

Resources

About