Classification of Drivers' Workload Using Physiological Signals in Conditional Automation.

Quentin Meteier,Elena Mugellini,Omar Abou Khaled,Marino Widmer,Andreas Sonderegger,Marine Capallera,Simon Ruffieux,Leonardo Angelini

doi:10.3389/fpsyg.2021.596038

Quentin Meteier, Elena Mugellini + Show 6 more

Open Access

https://doi.org/10.3389/fpsyg.2021.596038

Copy DOI

Abstract

The use of automation in cars is increasing. In future vehicles, drivers will no longer be in charge of the main driving task and may be allowed to perform a secondary task. However, they might be requested to regain control of the car if a hazardous situation occurs (i.e., conditionally automated driving). Performing a secondary task might increase drivers' mental workload and consequently decrease the takeover performance if the workload level exceeds a certain threshold. Knowledge about the driver's mental state might hence be useful for increasing safety in conditionally automated vehicles. Measuring drivers' workload continuously is essential to support the driver and hence limit the number of accidents in takeover situations. This goal can be achieved using machine learning techniques to evaluate and classify the drivers' workload in real-time. To evaluate the usefulness of physiological data as an indicator for workload in conditionally automated driving, three physiological signals from 90 subjects were collected during 25 min of automated driving in a fixed-base simulator. Half of the participants performed a verbal cognitive task to induce mental workload while the other half only had to monitor the environment of the car. Three classifiers, sensor fusion and levels of data segmentation were compared. Results show that the best model was able to successfully classify the condition of the driver with an accuracy of 95%. In some cases, the model benefited from sensors' fusion. Increasing the segmentation level (e.g., size of the time window to compute physiological indicators) increased the performance of the model for windows smaller than 4 min, but decreased for windows larger than 4 min. In conclusion, the study showed that a high level of drivers' mental workload can be accurately detected while driving in conditional automation based on 4-min recordings of respiration and skin conductance.

Highlights

According to the National Highway Traffic Safety Administration (NHTSA), 2,935 fatal crashes occurred on U.S roadways due to driver’s distraction in 2017
On the basis of decisions taken by the authorities concerned, drivers could be allowed to engage in a Non-DrivingRelated Task (NDRT) during periods of conditionally automated driving
The distraction induced by performing a NDRT using another sensory channel might increase the mental workload (MWL; Mehler et al, 2009)

Summary

Introduction

According to the National Highway Traffic Safety Administration (NHTSA), 2,935 fatal crashes occurred on U.S roadways due to driver’s distraction in 2017. Performing a secondary task while driving is one cause that increases the risk to have an accident, among other factors such as fatigue, mood or demanding driving conditions The latter lead to hazardous drivers states as named by Darzi et al (2018). According to the Society of Automotive Engineers (SAE) classification (SAE, 2018), the generation of vehicles that will emerge on our roads will be conditionally automated cars, corresponding to Level 3 of the SAE taxonomy At this automation level, the driver will no longer be in charge of the main driving task, neither monitoring the environment. The engagement of drivers in a NDRT would distract them from the supervision of the environment for which they are responsible They could be distracted visually, orally, cognitively, or biomechanically (Pettitt et al, 2005). The distraction induced by performing a NDRT using another sensory channel might increase the mental workload (MWL; Mehler et al, 2009)

Results

Discussion

Conclusion