Determination of dangerous driving behavior based on the use of information from wearable electronic devices

Monitoring the driver's behavior in the cabin of a vehicle is an urgent task for modern automated driver support systems (Advanced Driver Assistance Systems), which belong to the class of active safety systems. Existing research and solutions in the field of modern driver assistance systems are more focused on the use of electronic devices in the form of video cameras, lasers, and radars that provide measurement information about the driver in the cabin. However, the use of wearable electronic devices that measure the heart rate, electrocardiogram, user movements, and other indicators, allows one to obtain results about the driver's dangerous behavior more accurately and reliably. The paper proposes an approach to detecting dangerous states in the driver's behavior in the cabin of a vehicle based on the use of information from wearable electronic devices. The study shows that it is sufficient to use heart rate measurements passed from wearable electronic devices to detect dangerous states, such as aggression and stress. The developed mobile application on the Android platform allows detecting signs of aggression and stress in the driver's behavior using data obtained from sensors of wearable electronic devices. When the driver shows dangerous behavior in the cabin, the mobile application warns the driver by vibrating the wearable electronic device and an audio signal played by the smartphone. The developed approach is tested on a data set collected in real driving conditions on public roads in the city and on country roads in various driving conditions. Detection of signs of aggression and stress in the driver's behavior allows one to supplement information about the driver, and thereby improve the effectiveness of driver monitoring systems in the cabin of the vehicle, aimed at preventing and reducing the risk of road accidents and improving the skills of road users. The proposed approach can be used in combination with other technologies for monitoring driver behavior when building an intelligent driver support system.