Vehicular Networking for Intelligent and Autonomous Traffic Management

Despite the push for using mass transit, the number of vehicles on the road is growing at a steady rate and traffic congestion has become a daily problem that most people suffer. This not only impacts the productivity of the population but also poses a safety risk. Most of the technologies for intelligent highways focus on safety measures and increased driver awareness, and expect a centralized management for the traffic flow. We present a new approach for enabling autonomous and adaptive traffic management through vehicular networks. By allowing data exchange between vehicles about route choices, congestion and traffic alerts, a vehicle makes a decision on the best course of action. Unlike centralized schemes that provide recommendations, our VANET-based Autonomous Management (VAM) approach factors in the destination and routes of nearby vehicles in deciding on whether rerouting is advisable. In addition, VAM leverages the presence of smart traffic lights and enables coordination between vehicles and light-controllers in order to ease congestion. The collective effect of all vehicles will be an autonomous reshape of the traffic pattern based on their destinations and road conditions. To validate our approach we have developed a graphical tool that not only enables the collection of performance statistics but also allows visualizing the effect on traffic. The implementation also supports smart traffic lights and configurable roads. The simulation results demonstrate the advantage of VAM and there is up-to 40% increase in overall throughput for fully cooperative drivers.