Urban congestion and transportation inefficiencies pose significant challenges that require innovative solutions. Traditional static traffic signal control methods struggle to adapt to dynamic traffic conditions, leading to delays and increased pollution. Adaptive Traffic Control using Deep Reinforcement Learning (DRL) represents a promising approach to enhance traffic signal systems' performance by dynamically adjusting signal timings based on real-time traffic patterns. This paper introduces an adaptive traffic signal controller that minimizes delays and effectively manages heavy traffic volumes. The controller optimizes signal phases for efficient traffic flow and reduced congestion by using DRL to learn from and enhance its decision-making process. The system's performance and adaptability are assessed using metrics such as average speed, queue length, Number of vehicles, and waiting time. Deep Q-Network-based algorithm is introduced, featuring a custom Q-function, which is developed and compared against existing algorithms in diverse scenarios. This comparison showcases the new algorithm's superior performance and effectiveness in addressing dynamic traffic situations. This study advances urban traffic management by introducing an intelligent and adaptable traffic signal controller. This innovation has the potential to significantly improve transportation system efficiency and urban mobility.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Adaptive Traffic Control Using Deep Reinforcement Learning


    Contributors:
    N, Dhashyanth (author) / R, Hemchand (author) / R, Priyanga (author) / S, Soorya (author) / P, Sudheesh (author)


    Publication date :

    2024-12-13


    Size :

    1874349 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Adaptive Traffic Signal Control System Using Deep Reinforcement Learning

    Agrawal, Satyam / Sharma, Ritvij / Srivastava, Pankaj et al. | IEEE | 2024


    Reinforcement Learning for Adaptive Traffic Signal Control Using Deep Q-Networks

    Bohra, Asha Rani / Selvi, Thamarai / Vishal, V et al. | IEEE | 2025


    Deep Learning vs. Discrete Reinforcement Learning for Adaptive Traffic Signal Control

    Shabestary, Soheil Mohamad Alizadeh / Abdulhai, Baher | IEEE | 2018


    RA-TSC: Learning Adaptive Traffic Signal Control Strategy via Deep Reinforcement Learning

    Du, Yu / ShangGuan, Wei / Rong, Dingchao et al. | IEEE | 2019


    Adaptive traffic signal control system using composite reward architecture based deep reinforcement learning

    Jamil, Abu Rafe Md / Ganguly, Kishan Kumar / Nower, Naushin | Wiley | 2020

    Free access