Towards a Cooperative Intelligent System for Unpredictable and Predictable Road Hazard Detection

Abstract

Road hazards can cause dangerous accidents which lead to serious effects on human safety, destruction of vehicles and traffic flow disorder. Therefore, numerous systems based on Vehicular Ad hoc Networks (VANET) have been proposed to prevent this kind of accidents and enhance road users’ safety. Nevertheless, these systems suffer from some problems, which can reduce their performance. For example, some of the proposed systems are autonomous; they do not exploit VANET to cooperate. In addition, as these systems offer more and more features, they treat a large amount of data, but without storage in a database. This can lead to the problems of congestion or loss of data. Therefore, we propose in this paper a new system, entitled Cooperative Road Hazard Detection Persistent System (CopRoadHazDPS). This system is based on the use of (i) Vehicle-To-Vehicle (V2V) and Vehicle-To-Infrastructure (V2I) communications to promote cooperation between vehicles, infrastructures and the Control Center, and (ii) Real-Time DataBases (RTDB) to manage data in real-time, effectively and accurately. It ensures the road safety in the case of unpredictable and predictable road hazards. Once a vehicle or the Control Center identifies a road hazard, its Road Manager cooperates with the other components to analyze the situation and decides about the convenient actions to avoid accidents. Simulations of several driving scenarios in varied type of roads, within the Vehicles In Network Simulation (VEINS) framework, confirm that CopRoadHazDPS ensures safety and reduces data freshness transactions thanks to the concept of the Quality of Data.