Reinforcement Learning for Adaptive Traffic Signal Control at Urban Intersections

Ahmad Fauzi, Sari Handayani

Jurnal Teknologi dan Inovasi, Vol. 1 No. 2 (2026), pp. 81-92

Abstract

Fixed-time traffic signals cannot respond to fluctuating demand, causing long queues during peak hours. We train a deep reinforcement learning agent to choose signal phases at a four-arm intersection, using queue lengths and waiting times as the state and cumulative delay as the reward. The agent was trained and evaluated in a microscopic traffic simulation calibrated with two weeks of loop-detector counts. Compared with the existing fixed-time plan, the agent reduced average vehicle delay by 23% in the morning peak and 17% across the full day, without increasing pedestrian waiting time beyond the regulatory limit. Robustness tests with 20% higher demand still showed a 14% reduction in delay. The study outlines the steps needed before a field trial.

Keywords: reinforcement learning; traffic signal control; simulation; intelligent transportation; deep Q-network

1. Introduction

This section introduces the problem of intelligent transportation systems addressed in this study, reviews the main approaches reported in prior work, and states the research objectives.

2. Methods

We describe the data, the proposed approach, the baselines, and the evaluation protocol used to assess the results summarised in the abstract.

3. Results and Discussion

The results are presented and compared with the baselines. Limitations and practical implications for deployment are discussed.

4. Conclusion

The study shows that the proposed approach is effective for the stated objectives and outlines directions for future work.

References

Sutton, R. S., & Barto, A. G. (2018). Reinforcement Learning: An Introduction (2nd ed.). MIT Press.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539