Detecting Driver Fatigue based on the Driver´s Response Pattern and the Front View Environment of an Automobile

In the field of an automotive research, a method to monitor and to detect a drowsy or a drunken driver has been studied for many years. Previous research uses sensors such as an infrared camera for pupil detection or voice to detect fatigue. Even these approaches are able to detect driver¿s fatigue, however, these methods are not driver adaptable nor interactive with a outside driving situation. Unlike previous approach, we propose driver¿s fatigue detection system which uses the driver¿s pedal controlling pattern with respect to the driver¿s front view situation. The system uses a distance sensor on the frontend of the car so that it can capture an outside event. The entire sensor data are processed using a combination of Decision Tree learning algorithm and rule-based algorithm. The system does learning process at every startup of a car so that our system is capable to be adapted to each driver¿s driving style and behavior. Accordingly, we can be obtained the driver¿s fatigue level based on the response patterns.