Routing Strategies to Enhance Situational Awareness of Connected Vehicles in Unorganised Traffic Junctions

Vehicular Ad hoc Networks (VANETs) are the key to revolutionizing connected and Intelligent Transportation Systems (ITS) for the next generation. The South Asian/African countries typically have dense roads and unorganized traffic. However, this is seldom investigated. Junctions in these regions typically witness more than 50 vehicles making them accident-prone hot spots. Also, there is significant heterogeneity in vehicle type, physicality, and behavior. This paper mainly focuses on evaluating the strategies for unorganized traffic situations at four road junctions via Agent-based modeling. Specifically, it delves into investigating the lane-changing and overtaking behaviors of vehicles in such chaotic junctions. The performance is analyzed using NS3 and SUMO traffic simulators, for different routing protocols: Topology-based (AODV, DSDV, OLSR) and position-based (GPSR, PA-GPSR, MM-GPSR). Different metrics such as goodput, end-to-end delay, Packet Delivery Ratio (PDR), and Overhead are considered. Notably, the position-based protocols demonstrate significant performance in all metrics, even when the distance and range of vehicles increase in traffic junctions.