This paper introduces Visible Light Communication (VLC) as an integrated approach to improving traffic signal efficiency and vehicle trajectory management at urban intersections. By combining VLC localization services with learning-based traffic signal control, a multi-intersection traffic control system is proposed. VLC utilizes light communication between connected vehicles and infrastructure, enabling joint transmission and data collection via mobile optical receivers. Atmospheric conditions affecting communication quality are considered, with an analysis of outdoor coverage maps. The system aims to reduce waiting times for pedestrians and vehicles while enhancing overall traffic safety. Flexible and adaptive, it accommodates diverse traffic movements during multiple signal phases. Cooperative mechanisms, transmission ranges, and queue/response interactions balance traffic flow between intersections, improving road network performance. Evaluated using the SUMO urban mobility simulator, the multi-intersection scenario demonstrates reduced waiting and travel times for both vehicles and pedestrians. A reinforcement learning scheme, based on VLC queuing/response behaviors, optimally schedules traffic signals. Agents at each intersection control traffic lights using VLC-ready vehicles' communication, calculating strategies to enhance flow and communicate with each other for overall optimization. The decentralized and scalable nature of the proposed approach, particularly for multi-intersection scenarios, is discussed, showcasing its potential applicability in real-world traffic scenarios.
|