Fatigue driving detection method based on EEG analysis in low-voltage and hypoxia plateau environment

Driving fatigue is often an important cause of traffic accidents. Monitoring psychological parameters of driver to detect fatigue state is an effective approach to prevent traffic accident. In order to study driving fatigue state in low-voltage and hypoxia plateau environment, the research involved the aspects of subjective monitoring and objective monitoring by driver's real-time electroencephalogram (EEG) signal which was obtained by field driving fatigue test. Nonlinear and linear methods were used to analyze EEG signal in awake, critical, and fatigue three typical states. The EEGLAB in the MATLAB toolbox was used in nonlinear method to analyze the power spectral density map of θ, α, β wave in three typical states. The new EEG eigenvalues were collected, and the EEG power characteristic values were calculated to evaluate the trend of EEG signal in linear method. Combined nonlinear and linear methods with subjective data analysis, the energy characteristic of (α + θ)/β and (α + β)/θ were recommended as the indicator to evaluate driving fatigue characteristics in low-voltage and hypoxia plateau environment. This study provided foundation of theory and examination for the design of driving fatigue warning device in low-voltage and hypoxia plateau environment, which is of great significance for the development of driver fatigue detection system.