Driver fatigue is an important cause of traffic accidents and the detection of fatigue driving has been a hot issue in automobile active safety during the past decades. The purpose of this study is to develop a novel method to detect fatigue driving based on electroencephalogram (EEG). The volunteer is asked to perform simulated driving tasks under different mental state while EEG signals are acquired simultaneously from six electrodes at central, parietal and occipital lobe, including C3, C4, P3, P4, O1 and O2. Due to the non-linearity of human brain responses, correlation dimension is estimated with G-P algorithm to quantify the collected EEGs. Statistical analysis reveals significant decreases from awake to fatigue state of the correlation dimension for all the channels across 5 subjects (awake state: 3.87±0.13; fatigue state: 2.76±0.34; p < 0.05, paired t-test), which indicates that the correlation dimension is a promising parameter in detecting fatigue driving with EEGs. (Author/publisher)
Samenvatting