Enhancing Traffic Management through Visible Light Communication-Driven Signaling and Cooperative Trajectories

Abstract

Visible Light Communication (VLC) is a promising solution proposed for optimizing traffic signals and vehicle trajectories at urban intersections. This approach utilizes light communication between connected vehicles (CVs) and infrastructure to enable coordinated traffic interactions. By leveraging streetlamps, intersection signals, and headlights, VLC facilitates the transmission of information between CVs and the infrastructure. The system is designed to be flexible and adaptive, accommodating diverse traffic movements across multiple signal phases. To evaluate the effectiveness of VLC, simulations are conducted using the SUMO urban mobility simulator. These simulations generate traffic flows and incorporate VLC mechanisms and relative pose concepts for queueing, requesting, and responding to interactions. To dynamically control traffic flows and alleviate congestion during peak hours, a deep reinforcement learning algorithm is employed. This algorithm optimizes traffic by utilizing both Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications. Comparisons are made between the traditional trajectory and signal optimization techniques. The results demonstrate the benefits of VLC in terms of throughput, delay, and the reduction of vehicle stops. In conclusion, VLC presents an integrated approach that harnesses light communication to optimize traffic signals and vehicle trajectories at urban intersections. Through simulations and comparisons, VLC proves its effectiveness in enhancing traffic efficiency and reducing congestion, offering promising insights for future urban traffic management systems.

Keywords

• visible light communication
• optical sensors
• cooperative traffic control
• connected vehicles
• deep reinforcement learning sumo simulation