Visible Light Communication (VLC) offers an innovative solution for vehicular communication, combining illumination and data transmission seamlessly within existing infrastructure. This paper presents a novel traffic management system that integrates VLC with Artificial Intelligence (AI) to enhance safety and reduce delays for pedestrians and vehicles. By leveraging reinforcement learning (RL), the system adapts traffic signals dynamically based on real-time data, achieving significant improvements in traffic flow and efficiency. The proposed approach employs Deep Reinforcement Learning (DRL), with AI agents managing individual intersections or entire networks. VLC’s rapid data exchange capabilities enable real-time queue/request/response interactions, optimizing signal operations. Simulations conducted in the SUMO traffic simulator confirm the system’s effectiveness, demonstrating reduced waiting times and improved travel experiences compared to traditional methods. A major challenge addressed is the scalability of traffic signal coordination across multiple intersections. The distributed RL approach facilitates semi-independent yet cooperative signal control, accommodating both pedestrian and vehicle dynamics for optimal traffic management. This solution exhibits adaptability across various scenarios, positioning it as a robust framework for urban mobility. Aligned with CyberSecPro's mission, this research emphasizes the integration of cybersecurity in smart city infrastructures. The VLC-based system not only optimizes traffic management but also ensures secure, privacy-respecting data exchanges, advancing safe and efficient innovations in urban traffic systems.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Enhancing traffic flow with visible light communication: a deep reinforcement learning approach


    Contributors:

    Conference:

    Next-Generation Optical Communication: Components, Sub-Systems, and Systems XIV ; 2025 ; San Francisco, California, United States


    Published in:

    Proc. SPIE ; 13374


    Publication date :

    2025-03-19





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Enhancing Urban Traffic Management with Visible Light Communication and Reinforcement Learning

    Galvão Gonçalo / Vieira Manuel Augusto / Vieira Manuela et al. | DOAJ | 2024

    Free access

    Deep reinforcement learning traffic light control method

    KONG YAN / LI YING / CHIH-CHAO YANG | European Patent Office | 2024

    Free access

    Multimodal Traffic Light Control with Connected Vehicles: A Deep Reinforcement Learning Approach

    Zhou, Runhao / Nousch, Tobias / Adam, Django et al. | IEEE | 2023



    Smart Traffic Light Controller using Deep Reinforcement Learning

    Abhishek, A / Nayak, Prathvik / Hegde, Krishna P et al. | IEEE | 2022