Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{189855,
        author = {Shravani Sharad Patil and Gayatri Vijay Chaudhari and khushee Bhanudas Bagal and Spandan Pravin Patil},
        title = {Multi-Sensor Deep Q-Network (MSF-DQN) Real Time Adaptive Traffic Signal Control},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {8},
        pages = {241-253},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189855},
        abstract = {The Environmental and Economic Consequences of Urban Traffic Congestion Are a Significant Problem for Cities, And Traditional Fixed-Time and Actuated Controllers Only Make These Problems Worse.1 In This Paper, An Innovative Adaptive Traffic Signal Control System Based on A Multi-Sensor Fusion Deep Q Network (MSF-DQN) Called the Automatic Traffic Signal Information System (ATSIS) Is Introduced. The Suggested MSF-DQN Agent Has Stronger State Representation Than Any Previous Work Due To The Way It Combines Data From Two Different Sensing Modules: (1) A Vision-Based Mod- ule That Uses YOLOv4 For Vehicle Detection And Real-Time Queue Length Estimation, And (2) A High-Reliability Presence Detection Module Using Ultrasonic And Infrared Sensors At The Stop Line To Confirm Vehicles; These Types Of Combinations Provide Strengths That Are Very Important When Dealing With The Diversity And Non-Lane Based Nature Of India’s Traffic Conditions. The Performance of the Proposed MSF-DQN Model Is Rigorously Tested in A Variety of Traffic Scenarios Using the SUMO (Simulation of Urban Mobility) Microscopic Traffic Simulation Environment Low Volume, Dynamic Peak Hour, And Full Conditions Compared to Standard Fixed-Time (FT) And Actuated Control (AC) Systems. The Results Show That the Proposed MSF-DQN Significantly Reduces the Average Delay of Vehicles and The Average Queue Length, Especially Under Busy and Changing Conditions. Therefore, This Shows That the DRL Agent Works as Expected and The Multi Sensor Fusion Method Used Is Effective.},
        keywords = {Sensor Fusion, Deep Q-Network (DQN), Adap- tive Traffic Signal Control, Intelligent Transportation Systems (ITS), YOLO, Indian Traffic, SUMO.},
        month = {January},
        }