In partially automated cars, it is vital to understand the driver state, especially the driver's cognitive load. This might indicate whether the driver is alert or distracted, and if the car can safely transfer control of driving. In order to better understand the relationship between cognitive load and the driver performance in a partially autonomous vehicle, functional near infrared spectroscopy (fNIRS) measures were employed to study the activation of the prefrontal cortex of drivers in a simulated environment. We studied a total of 14 participants while they drove a partially autonomous car and performed common secondary tasks. We observed that when participants were asked to monitor the driving of an autonomous car they had low cognitive load compared to when the same participants were asked to perform a secondary reading or video watching task on a brought in device. This observation was in line with the increased drowsy behavior observed during intervals of autonomous system monitoring in previous studies. Results demonstrate that fNIRS signals from prefrontal cortex indicate additional cognitive load during manual driving compared to autonomous. Such brain function metrics could be used with minimally intrusive and low cost sensors to enable real-time assessment of driver state in future autonomous vehicles to improve safety and efficacy of transfer of control.
Monitoring driver cognitive load using functional near infrared spectroscopy in partially autonomous cars
2016-06-01
674589 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
TRIGGERING AUTONOMOUS CONTROL BASED ON DRIVER COGNITIVE LOAD
Europäisches Patentamt | 2020
|Functional Brain Imaging of Train Driver by Functional Near-Infrared Spectroscopy (FNIRS)
British Library Conference Proceedings | 2006
|A Multi-sensor Autonomous Integrity Monitoring Approach for Railway and Driver-less Cars
British Library Conference Proceedings | 2018
|