Drowsy driving is a significant contributor to death and injury crashes on our nation’s highways accounting for more than 80000 crashes and 850 fatalities per year. Recent research using data from the 100-car study found that drowsy driving contributed to 22% to 24% of crashes and near-crashes observed. This paper describes an approach that detects impairment from drowsiness in real time using inexpensive vehicle-based sensors to detect drowsiness-related changes in drivers’ behavior.

Data were collected on the National Advanced Driving Simulator from 72 volunteer drivers. Three age groups (21-34, 38-51, and 55-68 years of age) drove through representative situations on three types of roadways (urban, freeway, and rural) at three times of day (9 am-1 pm, 10 pm-2 am, and 2 am – 4 am) representing different levels of drowsiness.

Driving data indicated that a complex relationship exists wherein driving performance improves with low levels of drowsiness in the early night session before degrading in the late night session. This study demonstrates the feasibility of detecting drowsiness with vehicle-based sensors. Results show that alcohol and drowsiness impairment do not allow for a single algorithm to detect both types of impairment;​ however a single algorithm approach with different training data for the different types of impairment may be successful. To detect impairment due to either alcohol or drowsiness, a more complex approach is necessary where separate algorithms are combined to work with each other. These results suggest promise in a vehicle-based approach to detecting and differentiating multiple types of impairment.