Adaptive Traffic Flow Management Using YOLO, LSTM, and Q-Learning

DOI

10.2991/978-94-6239-693-7_60How to use a DOI?

Keywords

Traffic Management; Intelligent Transportation Systems; YOLO; LSTM; Reinforcement Learning; Q-learning

Abstract

Traffic congestion continues to be a significant problem in Indian cities which grow quickly because of multiple vehicle types and unpredictable traffic patterns and irregular lane usage. Traditional traffic signal operations that depend on fixed schedules and sensors fail to adjust properly to these changing conditions. The current system creates longer wait times which results in inefficient traffic flow. The research presents an AI based traffic signal control system which combines three modern methods through YOLO for vehicle detection and LSTM for traffic prediction and Q-learning for signal timing optimization. The YOLO system operates by using camera footage to detect and count moving vehicles while the LSTM model analyzes recent traffic patterns to predict how congested roads will become. The Q-learning agent uses these predictions to select the best green time distribution that will decrease traffic delays. The system achieved better results through tests with real traffic footage and simulated scenarios which demonstrated a 32-45% reduction in average waiting time and a 28-40% decrease in queue length compared to standard control systems. The system provides emergency vehicle priority and congestion control features which allow it to operate effectively at high-traffic urban intersections.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - P. Sandeep Yadav
AU  - L. Dharma Rao
AU  - G. Vinuja
PY  - 2026
DA  - 2026/06/16
TI  - Adaptive Traffic Flow Management Using YOLO, LSTM, and Q-Learning
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 607
EP  - 614
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_60
DO  - 10.2991/978-94-6239-693-7_60
ID  - Yadav2026
ER  -