Optimization of Intelligent Traffic Control Based on iot and Reinforcement Learning for Congestion Reduction in Smart Cities


Tri Aristi Saputri(1*); Budi Sutomo(2); Dimas Akbar Maulana(3); Hendika Purnomo(4);

(1) Universitas Dharma Wacana
(2) Universitas Dharma Wacana
(3) Universitas Dharma Wacana
(4) Universitas Dharma Wacana
(*) Corresponding Author

  

Abstract


Traffic congestion has become a major challenge in Indonesian urban areas due to rapid vehicle growth and the limited adaptability of conventional traffic signal control systems. Most existing Deep Reinforcement Learning (DRL)-based traffic signal control studies adopt a free-phase selection approach, which assumes full agent freedom in determining signal phases — an assumption fundamentally incompatible with fixed phase-sequence regulations in Indonesian urban infrastructure — and rely on synthetic traffic data that fails to represent motorcycle-dominated traffic conditions. Furthermore, existing DQN-based approaches treat all traffic density conditions uniformly, without utilizing IoT-derived density categories for context-aware decision-making. To address these gaps, this study proposes a manual phase rotation mechanism with constrained actions (15, 30, and 60 seconds) compatible with existing fixed-phase infrastructure without hardware modifications, real-world IoT CCTV data from four intersections in Metro City processed using the YOLOv11 model to generate Low, Medium, and High traffic density categories as a representative training foundation for Indonesian urban traffic conditions, and a category-based action bias mechanism that adjusts DQN Q-value estimates according to IoT-derived traffic density, enabling context-aware signal duration selection. The DQN agent interacts with the SUMO simulation environment through the TraCI interface, receiving real-time traffic states comprising vehicle count, queue length, waiting time, average speed, density category, and delta queue, and selecting optimal green signal durations based on an epsilon-greedy exploration strategy and experience replay mechanism over 1,100 training episodes. Training yielded a 39.2% improvement in total reward and a 6.6% reduction in average waiting time. The best-performing model, obtained at episode 1050, achieved an 8.6% reduction in average waiting time and an 11.7% increase in traffic throughput compared to the fixed-time baseline. These results demonstrate that the proposed framework contributes three concrete advances for adaptive traffic signal control, a constrained-action DQN that is fully compatible with real-world fixed-phase infrastructure, a real-world IoT CCTV dataset as a representative data foundation for Indonesian traffic conditions, and a category-based bias mechanism for context-aware control — collectively offering a deployable, infrastructure-compatible, and replicable solution for traffic authorities and local governments advancing the smart city agenda in Indonesia.


Keywords


IoT; SUMO; Reinforcement Learning; DQN; Smart Transportation

  
  

Full Text:

PDF
  

Article Metrics

Abstract view: 113 times
PDF view: 13 times 		     

Digital Object Identifier

doi  https://doi.org/10.33096/ilkom.v18i2.3260.320-332 		  

Cite

How to cite item

References


S. Damadam, M. Zourbakhsh, R. Javidan, and A. Faroughi, “An Intelligent IoT Based Traffic Light Management System: Deep Reinforcement Learning,” Smart Cities, vol. 5, no. 4, pp. 1293–1311, Sep 2022, doi: 10.3390/smartcities5040066.

J. Zang, P. Jiao, S. Liu, X. Zhang, G. Song, and L. Yu, “Identifying Traffic Congestion Patterns of Urban Road Network Based on Traffic Performance Index,” Sustainability, vol. 15, no. 2, pp. 948, Jan 2023, doi: 10.3390/su15020948.

Y. Wu, “Enhancing Urban Traffic Flow Through Fuzzy Logic-Based Signal Light Control Optimization:,” International Journal of e-Collaboration, vol. 20, no. 1, pp. 1–13, Nov 2024, doi: 10.4018/IJeC.358746.

J. Moreno-Malo, J.-L. Posadas-Yagüe, J. C. Cano, C. T. Calafate, J. A. Conejero, and J.-L. Poza-Lujan, “Improving traffic light systems using Deep Q-networks,” Expert Systems with Applications, vol. 252, pp. 124178, Oct 2024, doi: 10.1016/j.eswa.2024.124178.

R. Al-qudah, Y. Khamayseh, M. Aldwairi, and S. Khan, “The Smart in Smart Cities: A Framework for Image Classification Using Deep Learning,” Sensors, vol. 22, no. 12, pp. 4390, Jun 2022, doi: 10.3390/s22124390.

Y. Tian et al., “Traffic signal active control method for short-distance intersections,” PLoS ONE, vol. 20, no. 3, pp. e0319804, Mar 2025, doi: 10.1371/journal.pone.0319804.

S. Bouktif, A. Cheniki, A. Ouni, and H. El-Sayed, “Deep reinforcement learning for traffic signal control with consistent state and reward design approach,” Knowledge-Based Systems, vol. 267, pp. 110440, May 2023, doi: 10.1016/j.knosys.2023.110440.

R. Dai and Y. Li, “Double deep network-based traffic signal optimization method for isolated intersections,” DTS, vol. 5, no. 1, pp. 42–52, 2026, doi: 10.48130/dts-0026-0004.

Y. Lin, A. Xie, and X. Liu, “Autonomous Vehicle Decision and Control through Reinforcement Learning with Traffic Flow Randomization,” Machines, vol. 12, no. 4, pp. 264, Apr 2024, doi: 10.3390/machines12040264.

C. Cai and M. Wei, “Adaptive urban traffic signal control based on enhanced deep reinforcement learning,” Sci Rep, vol. 14, no. 1, pp. 14116, Jun 2024, doi: 10.1038/s41598-024-64885-w.

R. Muwardi et al., “YOLO-LIO: A Real-Time Enhanced Detection and Integrated Traffic Monitoring System for Road Vehicles,” Algorithms, vol. 19, no. 1, pp. 42, Jan 2026, doi: 10.3390/a19010042.

S. Mansouryar, C. Colombaroni, N. Isaenko, and G. Fusco, “Deep Reinforcement Learning Approach for Traffic Light Control and Transit Priority,” Future Transportation, vol. 5, no. 4, pp. 137, Oct 2025, doi: 10.3390/futuretransp5040137.

Y. Kong, Y. Li, and C.-H. Hsia, “MALight: A Deep Reinforcement Learning Traffic Light Control Algorithm with Pressure and Attentive Experience Replay,” JIT, vol. 25, no. 7, pp. 955–962, Dec 2024, doi: 10.70003/160792642024122507001.

S. M. M. R. Swapno et al., “A reinforcement learning approach for reducing traffic congestion using deep Q learning,” Sci Rep, vol. 14, no. 1, pp. 30452, Dec 2024, doi: 10.1038/s41598-024-75638-0.

I. A. Salenek, Y. A. Seliverstov, S. A. Seliverstov, and E. A. Sofronova, “Improving the quality of route generation in SUMO based on data from detectors using reinforcement learning,” CRM, vol. 16, no. 1, pp. 137–146, Feb 2024, doi: 10.20537/2076-7633-2024-16-1-137-146.

B. Muyombo and H. Vadapalli, “Enhancing Urban Traffic Flow: An AI-Driven Traffic Light Management System,” Procedia Computer Science, vol. 270, pp. 3848–3859, 2025, doi: 10.1016/j.procs.2025.09.510.

S. Baeva, N. Hinov, and P. Nakov, “Comparative Analysis of Some Methods and Algorithms for Traffic Optimization in Urban Environments Based on Maximum Flow and Deep Reinforcement Learning,” Mathematics, vol. 13, no. 14, pp. 2296, Jul 2025, doi: 10.3390/math13142296.

K. Khalid, A. Bouayad, and K. Fardousse, “Adaptive Traffic Lights Control System Using Deep Q-Network and Salp Swarm Optimization,” IEEE Access, vol. 13, pp. 114199–114207, 2025, doi: 10.1109/ACCESS.2025.3583656.

W. Zai and D. Yang, “Improved Deep Reinforcement Learning for Intelligent Traffic Signal Control Using ECA_LSTM Network,” Sustainability, vol. 15, no. 18, pp. 13668, Sep 2023, doi: 10.3390/su151813668.

P. Michailidis, I. Michailidis, C. R. Lazaridis, and E. Kosmatopoulos, “Traffic Signal Control via Reinforcement Learning: A Review on Applications and Innovations,” Infrastructures, vol. 10, no. 5, pp. 114, May 2025, doi: 10.3390/infrastructures10050114.

H. Zhang, Z. Fang, Y. Chen, H. Dai, Q. Jiang, and X. Zeng, “Traffic signal optimization control method based on attention mechanism updated weights double deep Q network,” Complex Intell. Syst., vol. 11, no. 5, pp. 217, May 2025, doi: 10.1007/s40747-025-01841-9.

H. Xue, P. Liu, and Z. Mu, “Research on Traffic Signal Control Algorithm Based on Deep Reinforcement Learning,” International Journal of Advanced Network, Monitoring and Controls, vol. 10, no. 3, pp. 72–80, Sep 2025, doi: 10.2478/ijanmc-2025-0027.

K. Cao et al., “Optimization Control of Adaptive Traffic Signal with Deep Reinforcement Learning,” Electronics, vol. 13, no. 1, pp. 198, Jan 2024, doi: 10.3390/electronics13010198.

R. A. Velasquez Ortiz, M. E. Lárraga Ramírez, L. A. Alvarez-Icaza, and H. A. Guzmán Gutiérrez, “Adaptive Traffic Signal Control Using Deep Reinforcement Learning with Noise Injection,” Applied Sciences, vol. 16, no. 8, pp. 3833, Apr 2026, doi: 10.3390/app16083833.

H. Meess, J. Gerner, D. Hein, S. Schmidtner, G. Elger, and K. Bogenberger, “First steps towards real-world traffic signal control optimisation by reinforcement learning,” Journal of Simulation, vol. 18, no. 6, pp. 957–972, Nov 2024, doi: 10.1080/17477778.2024.2364715.

J. Lv, Z. Wang, and J. Ma, “Carbon Emission Reduction in Traffic Control: A Signal Timing Optimization Method Based on Rainbow DQN,” Applied Sciences, vol. 15, no. 3, pp. 1101, Jan 2025, doi: 10.3390/app15031101.

Z. Sun et al., “Joint control of traffic signal phase sequence and timing: a deep reinforcement learning method,” Digital Transportation and Safety, vol. 4, no. 2, pp. 118–126, 2025, doi: 10.48130/dts-0025-0008.

S. Sun and M. Liu, “Optimisation of Lane‐Level Dynamic Traffic Control Strategy Based on Bidirectional Adaptive Gated Graph Convolutional Network and Deep Reinforcement Learning,” IET Intelligent Trans Sys, vol. 19, no. 1, pp. e70055, Jan 2025, doi: 10.1049/itr2.70055.

S. R. M. Savithramma and R. Sumathi, “Intelligent traffic signal controller for heterogeneous traffic using reinforcement learning,” Green Energy and Intelligent Transportation, vol. 2, no. 6, pp. 100124, Dec 2023, doi: 10.1016/j.geits.2023.100124.

W. Bouzi, S. Bentaieb, and A. Ouamri, “Optimizing Traffic Light Control using Enhanced DQN: Minimizing Waiting Time for Regular and Emergency Vehicles,” Transport and Telecommunication Journal, vol. 26, no. 3, pp. 266–275, Apr 2025, doi: 10.2478/ttj-2025-0020.

C. Wu, I. Kim, and Z. Ma, “Deep Reinforcement Learning Based Traffic Signal Control: A Comparative Analysis,” Procedia Computer Science, vol. 220, pp. 275–282, 2023, doi: 10.1016/j.procs.2023.03.036.

D. C. Khrisne, “Optimizing traffic lights at unbalanced intersections using deep reinforcement learning,” IAES International Journal of Artificial Intelligence, vol. 14, no. 4, pp. 2991–3002, Aug. 2025.

H.-Y. Lin and M.-Y. Tsai, “Traffic light detection for nighttime driving with log-polar transform incorporated learning,” Results in Engineering, vol. 25, pp. 103783, Mar 2025, doi: 10.1016/j.rineng.2024.103783.


Refbacks

  • There are currently no refbacks.


Copyright (c) 2026 Tri Aristi Saputri, Budi Sutomo, Dimas Akbar Maulana, Hendika Purnomo

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.