A new drowsy-driver detection system that tracks and classifies steering behavior is being developed at George Washington University. The system can detect the differences between drowsy and non-drowsy driver behaviors with 90% accuracy.
Background
National Highway Traffic Safety Administration’s Fatality Analysis Reporting System (FARS) tracks crash statistics
Fatigue involved in 10-40% of crashes on long motorways, 15% of fatal single vehicle truck crashes
System uses artificial neural networks which attempt to mimic brain’s neuron processing
Functions mimicked include pattern recognition, signal processing, and control systems