Enhanced ALIVE Mind Controller and Machine Learning to Detect Drowsiness While Driving

Thousands of road accidents occur each year, one of the main causes being drowsiness at the wheel. Therefore, this paper proposes an enhanced scheme called ALIVE Mind to detect drowsy drivers and take action on their car using Machine Learning and EEG. For that matter, we built and designed a circuit board called AMC 2.0 that allows a simple EEG headset to read brain waves from a driver and send this data to a computer via Bluetooth. Those signals are then saved and analyzed by a deep neural network to find if the driver is drowsy or is about to sleep. To evaluate our solution, we conducted simulations and collected brain signals of a subject driver while on a driving simulator. With those values, we were able to train a model to detect whether the subject was drowsy or awake. Finally, when the system detects that the driver is in a fatigued state, it can take control of the car to park it in a safe place. Preliminary results show the effectiveness of the ALIVE Mind project and how it outperforms previous works in terms of minimizing computational needs and improving the prototype's convenience and suitability for car constructors.