Fatigue of car drivers - detection and classification based on the experiments on car simulators

The paper discusses the problem of drowsiness and phenomenon of microsleeps of healthy drivers. It describes our experiments done on our driving simulators. First there is a brief description of simulator which was used for the experiments, its capabilities and necessary changes which we did so that it fits best the needs of such experiments. The paper presents requirements on the overall design of experiments performed on driving simulators. It describes requirements on the design of the testing tracks and our experiences with measurements. The results of classification based on EEG, video record expert analysis, steering wheel correction movements, lateral position and average velocity change trends and weaving and drive out from the lane are presented. Finally discussion on analysis methods used for investigation in driver's drowsiness is done.