IoV based intelligent vehicle tracker using FoG computing with supervised machine learning techniques

Abstract

IoVs (Internet of Vehicles) are prominently being deployed for tracking the moving vehicles on the roads. Exclusively, the combination of IoVs and Fog computing is very useful to reduce the rate of road-accidents by reducing the chances of collisions of moving vehicles in real-time. The most prominent factors behind road-accidents are the high-speed driving, mixing the drinking and driving and ignorance to the traffic rules. To reduce the chances of road accidents, the accurate location, the route information, the speed of vehicle and the alcohol degree consumed are to be monitored actively and constantly. The real-time data with these parameters, is processed with IoVs architecture and fog computing based high computational powers. Using machine learning algorithms, the Vehicles moving on the road are classified as risky (accident-prone) or non-risky (non-accident-prone). Now, the vehicle which comes under the class of “risky” one, in real time are saved by sending safety alarms. Multiple machine learning algorithms are implemented to find the most accurate model of detection of the accident. Overall, contribution of this work is to reduce the number of road accidents.

Subject Classification:

• 68T45
• 18B20

Keywords:

• Global positioning system (GPS)
• Internet of Vehicles (IoV)
• Internet of Things (IoT)
• Machine learning
• Vehicle tracking