A Robust Drowsiness Detection Method based on Vehicle and Driver Vital Data

Abstract

Driver drowsiness is one of the main causes of fatal traffic accidents. Current driver assistance systems often use parameters related to driving behavior for detecting drowsiness. However, the ongoing automation of the driving task diminishes the availability of driving behavior parameters, therefore reducing the scope of such detection methods. The driver’s role as the sole operator changes; the driver must supplement, supervise or serve as a fallback part of a highly assisted/automated system. Reliably monitoring the driver’s state, especially the risk factor drowsiness, becomes more and more important for future automated driver systems. Numerous approaches, utilizing vehicle-based, behavioral and physiological based metrics, exist. This paper summarizes and discusses prevailing research questions related to drowsiness modeling and detection within the automotive context. Focus is placed on the utilization of driver vital data measured by wearable and other in-car sensors.