The project is supported by VZ MSM 210000024, MD 802/210/108. Driver’s Micro-sleeps Detection Using Virtual Simulation Petr Bouchner Department of Control Engineering and Telematics Joint Laboratory of System Reliability Czech Technical University, Prague Faculty of Transportation Sciences Konviktská 20, 110 00 Prague, Czech Republic CZECH REPUBLIC Abstract: - This paper introduces the cooperation of a virtual car simulator and EEG measurements to test a human driver’s behavior in demanding situations. After a short explanation of the main principles and tasks of EEG measurements, basic concepts of our experiments are presented. The following part is devoted to problems and solutions concerning the physical model, graphical and other aspects of our simulator. At the end of the article the various measuring procedures are presented. Key-Words: - Driving simulation, Driving safety, EEG, Attention decrease, Driving in virtual environments 1 Introduction Majority of today’s available technical devices cannot operate without the help of a human operator. The operator can be very stressed and his job needs constant vigilance [1] [3]. In the Joint Laboratory of System Reliability, we work on projects dedicated to the analysis and prevention of the operator attention decrease and micro-sleep. One of the very promising methods that can be successfully used for micro-sleep detection and prediction, involves the analysis of driver’s EEG signal. It is possible to say that we can (at least manually) detect from the EEG the actual driver’s mental state and derive if he is about to fall into micro-sleep. The aim of this article is to inform about the development, structure and problems associated with the setup of the simulation device. (The particular results are presented here in other articles from our group.) 2 EEG Measurements The electric signals arise from an activity of neurons of the thalamus and cortex. A normal EEG signal is quasi- periodic and has approximately a sinusoidal shape. It is usually differentiated into [2]: • Delta - (0.5 - 4 Hz) - appears in deep sleep. It is also typical for analytical thinking. Occurrence during adult's vigilance is pathological. Its amplitude is usually between 10 and 200. • Theta - (4 - 8 Hz) - can be found together with delta activity in certain phases of sleep. Theta activity also increases during psycho-tests, even with open eyes. • Alpha - (8 - 13 Hz) – It is most apparent with closed eyes. It is damped by intellectual activity and opened eyes. Its amplitude is usually between 30 and 70. • Beta - (13 - 30 Hz) - is typical for uneasiness. The amplitude is up to 30. The maximum of beta activity is in the frontal part of the brain. 2.1 Measuring of Probands (Testing Persons) The original approach which we have used until now, was preformed with closed eyes. The proband was sitting in front of a screen and watching (in major time hearing) a non-interactive movie. As a reference of EEG signal a reaction time to an acoustic impulse was used [4] (Pushing a button near a proband’s finger.). During the measurement, the proband was falling to sleep as his eyes closed. According to the reaction time we divided the proband’s state of mind into three groups. • Vigility - reaction time within 150 and 400 ms • Relaxation - reaction time within 400 and 800 ms • Somnolence - reaction time over 800 ms to the time when the proband stops reacting at all. All above described states have their characteristic features. The most important is the disappearing of alpha activity and a shift of all the spectra into the region of delta waves as the reaction time increases [5]. 2.2 Measurement with 3D simulator The approach with non-interactive scenes had several disadvantages. The measurements should be based on the simulation of the work of a system operator. This condition is hard to fulfill if the operator cannot interact. Another problem was that the measurement was done with closed eyes and the only stimulation was of acoustic origin. However, the drivers during the ride