Based On Eeg Analysis Studies Of Grassland Road Driver Fatigue Level Reference Rhreshold

Recent years, with the continuous development of grassland people’s living standards and grassland tourism, the penetration of vehicles improve continually in grassland region, the density of road network and highway mileage increase year by year, resulting in an increase in traffic accidents every year in grassland region, where most of the accidents are caused by driver fatigue. Hence, promptly determining driver fatigue and effectively preventing the occurrence of fatigue driving, are of vital importance to reduce accidents and improve the grassland road traffic safety. It is believed that EEG is a "gold standard" for driver fatigue reflection, but many scholars in the study only investigated the differences in the driver awake and fatigue EEG ratio value s, while less report is focused on these two states EEG reference domains. To protect the driver’s traffic safety and determine the driver’s driving state during the driving process, accurate driver EEG reference threshold is needed. Therefore, in this paper, the driver’s EEG reference threshold of grassland road was studied to reduce and prevent grassland road accidents caused by driving fatigue.The collected EEG in this paper during the driving process on grassland road was regarded as the research object, paired T test method was employed on ratio value of α/β and (α+θ)/β of the driver’s EEG, thus the time quantum of the driver in both awake and fatigue state was obtained, and subjective questionnaire was used to verify the accuracy of awake and fatigue period. And then combining the thought of cumulative contribution rate of principal component analysis, a simple and accurate calculation method of reference domain was proposed, based on the above study, the reference threshold of EEG of awake and fatigue state was finally obtained. The result demonstrated that this algorithm can accurate and fast determine the reference domain of driver’s EEG during the driving process, has general applicability in the calculation of the driver’s driving behavior of each index, and is conducive to later study of driver fatigue detection. As a result, the following conclusions were obtained:(1) Combining the thought of cumulative contribution rate of principal component analysis, a calculation method which is suitable for calculation of threshold algorithm was proposed.(2) Using the AcqKnowledge software to analyze of brain waves,it can be found that the ratio value of and (α+θ)/β of the driver’s EEG increases with the increase of driving fatigue.(3) On grassland road, the fatigue state of driver occurs at about20-30min.(4) The reference threshold of a in EEG at awake state is9-13.08Hz, p is16.23-19.28Hz,(?) is4.73-5.07Hz; meanwhile, the reference threshold of a at fatigue state is7.99-9Hz,β is13.91-16.23Hz,(?) is5.17-5.41Hz.