Recent studies reveal that driving without sufficient sleep or anger situations would increase the risk of road traffic accident. Driving fatigue or anger has been considered as a significant risk factor in transportation accidents, and the development of the human cognitive state based on electroencephalogram (EEG) has become a major focus in the field of driving safety for many drivers. However, it faces portable and real-time problems on its practical application. This study uses MindWave to collect the Attention and Eye Blink Detection EEG from the left prefrontal lobe of the subject, and uses the relation between Attention and Eye Blink Detection EEG when the subject is in the state of concentration, relaxation, fatigue, sleep and anger being measured first. A new method for driving fatigue and anger detection based on the correlation coefficient between drivers Attention and Eye Blink Detection EEG is proposed. Meanwhile, the Matlab and Arduino algorithm is ThinkGear Library that introduced to classify the correlation coefficient between the drivers Attention and Eye Blink Detection EEG, so as to detect driving fatigue or anger and alert. Lastly, the software running on an Arduino and Matlab is developed based on the above technologies, and the experiment proves that it has non-invasive and real-time advantages. As a result, the system is needed to give an warning for driver when he/she feels drowsy or anger to take a rest. With using brain wave censor hopefully this application can give an early warning and real time warning for driver. According to National Highway Traffic Safety Administration, at least, 1,500 people die in crashes related to fatigued and anger drivers in the United States of America each year. At least 40,000 people are injured in drowsy and anger driver crashes in more than 100,000 crashes each year. According to Automaker like Ford, Volvo, and others respond to this issue by developing a system to give alert when the abnormal driving behaviours detected. But this feature is only available on the top-of-the-line cars or the new one. This feature cannot be added to a model for a custom features because many components in the system will need to be replaced and it must be done at the manufacturer site. Many efforts have been done to provide the solution for this problem. The first approach is using imaging to monitor driver’s condition. Ranging from driver’s pupils detection up to facial images analysis with neural network based algorithm for measuring the eye opening and closing. Another approach is using sensor based in-vehicle system, to detect driver’s drowsiness and anger. The drowsy and anger detection systems as mentioned before are relatively complex and have strong dependencies on some on- board sensors. Instead of implementing on-board sensor, this work focused on utilizing sensor which can have a direct contact with driver to detect drowsiness and anger. For that reason, an independent and portable system is developed to detect drowsiness and anger based on Electroencephalography (EEG) approach. This study is conducted to overcome those problem and forms more reliable solution. NeuroSky MindWave is used as a probe to detect the brain wave pattern of the user. This system can be used in various vehicles and can be moved from one to another. Library called ThinkGear is built based on Matlab and Arduino IDE using C++ and can be installed in Arduino UNO and make some analysis with Matlab Simulation. This device can automatically get data from MindWave, process it and give suggestion or alert to the users when the drowsy or anger state detected with attention value and eye blink detection in every minute. This system works in every minutes with Arduino and alert to driver of any anger or drowsiness situation. So, it contribute safety driving. (Author/publisher)
Abstract