Comprehensive evaluation method for plateau driving fatigue based on psychophysiological indicators

Abstract

To investigate the effects of plateau environments on driving fatigue, heart rate and electroencephalogram(EEG)signals were chosen as indicators to characterize driving fatigue. The study analyzed the variation in these indicators as drivers transitioned into fatigued stages. By examining the sample entropy of EEG signals and the heart rate variation coefficient, a complex indicator of driving fatigue(CIDF)was established using principal component analysis to overcome the limitations of single-indicator methods. According to the CIDF values, the driving fatigue states in plateau areas were subdivided into three categories, including alertness, mild fatigue, and severe fatigue, by cluster analysis. Optimal binning determined thresholds for different driving fatigue states, which were validated through variance analysis. The results indicate that the CIDF values effectively distinguish the driving fatigue states of drivers in plateau areas. The CIDF thresholds for the alertness and the mild fatigue states are 0.34 and 0.50, respectively. A CIDF value greater than 0.50 indicates that the driver is in a severe fatigue state.