Driver Drowsiness Detection Based on Time Series Analysis of Steering Wheel Angular Velocity

Abstract

A novel driver drowsiness detection method based on time series analysis of the steering wheel angular velocity is proposed in this paper. Firstly, the steering behavior under the fatigue state is analyzed, followed by the determination of the temporal detection window, and then, the data series of the steering wheel angular velocity in the temporal detection window is selected as the detection feature. IF the detection feature satisfies the extent constraint and the variability constraint in the temporal window, a drowsiness state is detected accordingly. At last, experiment tests validate our method has good performance and could be well used in the real world.