Monitoring Driver's Alertness Based on the Driving Performance Estimation and the EEG Power Spectrum Analysis

Abstract

Preventing accidents caused by drowsiness behind the steering wheel is highly desirable but requires techniques for continuously estimating driver's abilities of perception, recognition and vehicle control abilities. This paper proposes methods for drowsiness estimation that combine the electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Results show that it is feasible to quantitatively monitor driver's alertness with concurrent changes in driving performance in a realistic driving simulator.

Impact of refreshing perception on mood, cognitive performance and brain oscillations: An exploratory study

“…Oscillatory activity in the beta or higher frequency ranges seems to play an important role in the functional organization of neuronal activity that underlies visual perception, attention and working memory (Gross et al, 2004;Tallon-Baudry, 2003). The possibility to predict performance based on EEG oscillatory activity is further supported by other data showing that oscillations below 20 Hz, especially those peaking between 7-12 and 16-20 Hz, are positively correlated with driving performances (Liang et al, 2005). From these EEG data, alpha and beta oscillations may account for an increase of antero-posterior synchrony with better performance, possibly reflecting the activation of a neural network involved in visual attention control.…”

Driving Fatigue and Performance among Occupational Drivers in Simulated Prolonged Driving

“…The use of driving simulator enabled us to simulate actual driving situation in a safe and controlled environment [Campagne et al, 2004;Liang et al, 2005;Otmani et al, 2005;Philip et al, 2003;Roge et al, 2004;Schier, 2000). The driving simulator allowed us to conduct the study of long and monotonous driving task with different kinds of road and climatologically event, which in real traffic scene, could possibly bring to severe accidents.…”

EEG-based prediction of driver's cognitive performance by deep convolutional neural network

“…For instance, independent component analysis (ICA) was applied to the EEG data and classifiers that used independent components (ICs) selected based on the prior knowledge were demonstrated in a test involving 10 subjects to provide good performance in predicting driver's fatigue [1][2]. Time-frequency EEG power spectrum was also demonstrated as useful features in [3] on data from 16 subjects. A combination of raw EEG and time-frequency power spectrum features was also investigated for data from 4 subjects in [4].…”

“…Description of the data and formulation of the problem EEG measurements were collected from three studies of driver's cognitive performance conducted in a virtual reality dynamic driving simulator [1][2][3], where the designs of all studies are the same. Thirty electrodes are used with a modified version of the international 10-20 system (standard electrode positioning nomenclature, American Encephalographic Association).…”