Evaluation of driver fatigue with multi-indicators based on artificial neural network

Three kinds of signals include head nodding movement, eyelid closure and driving time are demonstrated to be the factors to indicate driver fatigue. On the basis of these factors, five fatigue-based indicators are proposed in this study, which are integrated to establish the evaluation model of driver fatigue with artificial neural network. A total of 50 drivers are recruited to take part in the fatigue-oriented experiments on a driving simulator. The electroencephalogram (EEG), head nodding angle, eye-tracking signal, driving time-of-day and time-on-task are sampled during experiments. The EEG-based indicator is determined and used to group the sample data into alert and drowsy. The head nodding-based indicator includes mean value and dominant frequency of head nodding angle is proposed. The results show that the performance of the proposed evaluation model is better than that with sole head nodding-based or eyelid closure-based indicator.

[1]  T. Åkerstedt,et al.  Transport and industrial safety, how are they affected by sleepiness and sleep restriction? , 2006, Sleep medicine reviews.

[2]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[3]  Toshiki Yoshimine,et al.  Temporal dynamics of wakefulness during simulated driving , 2007 .

[4]  Evangelos Bekiaris,et al.  Using EEG spectral components to assess algorithms for detecting fatigue , 2009, Expert Syst. Appl..

[5]  S D Baulk,et al.  Chasing the silver bullet: measuring driver fatigue using simple and complex tasks. , 2008, Accident; analysis and prevention.

[6]  Luis M. Bergasa,et al.  Controlled inducement and measurement of drowsiness in a driving simulator , 2010 .

[7]  Miguel Ángel Sotelo,et al.  Real-time system for monitoring driver vigilance , 2004, Proceedings of the IEEE International Symposium on Industrial Electronics, 2005. ISIE 2005..

[8]  T. Sejnowski,et al.  Estimating alertness from the EEG power spectrum , 1997, IEEE Transactions on Biomedical Engineering.

[9]  B. Oken,et al.  Vigilance, alertness, or sustained attention: physiological basis and measurement , 2006, Clinical Neurophysiology.

[10]  Wei Li,et al.  Evaluation of driver fatigue on two channels of EEG data , 2012, Neuroscience Letters.

[11]  A. Schlögl,et al.  Artifact Processing in Computerized Analysis of Sleep EEG – A Review , 1999, Neuropsychobiology.

[12]  I A Basheer,et al.  Artificial neural networks: fundamentals, computing, design, and application. , 2000, Journal of microbiological methods.

[13]  M. Chung,et al.  Electroencephalographic study of drowsiness in simulated driving with sleep deprivation , 2005 .

[14]  W. Dement,et al.  Awareness of sleepiness and ability to predict sleep onset: can drivers avoid falling asleep at the wheel? , 2007, Sleep medicine.

[15]  Zhiwei Zhu,et al.  Real-time nonintrusive monitoring and prediction of driver fatigue , 2004, IEEE Transactions on Vehicular Technology.

[16]  A. Muzet,et al.  Correlation between driving errors and vigilance level: influence of the driver's age , 2004, Physiology & Behavior.

[17]  A. Craig,et al.  Electroencephalography Activity Associated with Driver Fatigue: Implications for a Fatigue Countermeasure Device , 2001 .