Active Search Mirror Checks Dataset (05-084)

Project Description: Driver impairment, due to drowsiness or fatigue, has a significant impact on the safety of all road users. Assessing an impairment such as driver drowsiness, through the use of vehicle-based technology, continues to be an area of interest. Both the initial detection, as well as continued monitoring, of driver drowsiness have been the emphasis of vehicle-based Driver Monitoring Systems (DMS). Particularly, in-vehicle eye tracking systems have been implemented, as a way of determining driver state. Specifically when hands-free driving assistance features are engaged, measures such driver’s percentage of eye closure (PERCLOS) are being used to determine driver drowsiness. However, one challenge of such a metric is its reliability; particularly with regard to false alarms (when a DMS indicates the driver is drowsy, but in fact is not). Therefore, the use of more gross-level driver behavioral based measures may serve as a way of crosschecking the assessments of a DMS. This work aims to mine an available dataset in order to examine driver search behavior, with the goal of identifying relationships between driver vigilance and drowsy driving. The hypothesis is that driver search behavior (e.g. mirror checks) degrades with increasing levels of drowsiness. If a reliable relationship is found between driver vigilance and state of drowsiness, the practical applications may be to incorporate this measure of driver search behavior into the “toolbox” of metrics for estimating driver drowsiness. Data Scope: The data are 135 epochs of drivers through varying PERCLOS Classifications. Timestamps indicate the beginning and ending of 5-minute epochs with glances to mirrors and over-the-shoulder areas. Data Specification: Please see Data Dictionary document below.