Smart City Gnosys

Smart city article details

Title	A New Traffic Signaling Model Based On Graph And Deep Reinforcement Learning; [Graf Ve Derin Pekiştirme Öǧrenme Tabanli Yeni Bir Trafik Sinyalizasyon Modeli]
ID_Doc	3158
Authors	Turan E.; Dandil B.; Avci E.
Year	2024
Published	Journal of the Faculty of Engineering and Architecture of Gazi University, 40, 1
DOI	http://dx.doi.org/10.17341/gazimmfd.1257860
Abstract	Topological structure and waiting times of vehicles at the intersection are shown as common causes of traffic congestion. Since the improvements in the topological structure can be realized as a result of long and costly projects, intersection signaling applications become an indispensable application area of smart cities. In intersection signaling applications, phase sequence and duration are calculated to ensure the maximum flow of vehicles per unit time, on the basis of intersection or throughout the network. Junction signaling optimization is a real-time realworld problem that is affected by many variable data. Therefore, many studies are still being carried out to develop the most efficient signaling method. In this study, an approach that optimizes the order and duration of the green phase is proposed to reduce the waiting times at the junction points across the network. This approach has been developed by transferring real-time vehicle data to the SUMO simulator according to the exact scale of city intersections on the real-world map. A new signaling approach named GDRL is proposed by combining graphbased phase duration and Deep Reinforcement Learning-based phase sequence estimation. In this approach, the phase sequence is calculated by the DRL method. The phase duration is calculated based on the maximum flowfinding method of the Ford-Fulkerson algorithm. The GDRL approach has been tested in the SUMO simulator by running parallel on consecutive intersections on the real map and using real data. It has been observed that the GDRL approach efficiently solves traffic congestion by reducing the queue length at intersections by 44%. © 2024 Gazi Universitesi. All rights reserved.
Author Keywords	Deep reinforcement learning; Max flow graph; SUMO simulator; Traffic signaling

Similar Articles

Id	Similarity	Authors	Title	Published
23735	0.907	Sattarzadeh A.R.; Pathirana P.N.	Enhancing Adaptive Traffic Control Systems With Deep Reinforcement Learning And Graphical Models	Proceedings - 2024 IEEE International Conference on Future Machine Learning and Data Science, FMLDS 2024 (2024)
43250	0.899	Sattarzadeh A.R.; Pathirana P.N.	Probabilistic Graph Models: A Key To Boosting Deep Reinforcement Learning In Urban Traffic Networks	2025 17th International Conference on Computer and Automation Engineering, ICCAE 2025 (2025)
25829	0.897	Thamaraiselvi K.; Bohra A.R.; Vishal V.; Sunkara P.S.; Sunku B.; Nityajignesh B.	Exploring Traffic Signal Control: A Comprehensive Survey On Reinforcement Learning Techniques	3rd IEEE International Conference on Industrial Electronics: Developments and Applications, ICIDeA 2025 (2025)
12771	0.896	Cao M.; Li V.O.K.; Shuai Q.	Book Your Green Wave: Exploiting Navigation Information For Intelligent Traffic Signal Control	IEEE Transactions on Vehicular Technology, 71, 8 (2022)
21429	0.888	Skoropad V.N.; Deđanski S.; Pantović V.; Injac Z.; Vujičić S.; Jovanović-Milenković M.; Jevtić B.; Lukić-Vujadinović V.; Vidojević D.; Bodolo I.	Dynamic Traffic Flow Optimization Using Reinforcement Learning And Predictive Analytics: A Sustainable Approach To Improving Urban Mobility In The City Of Belgrade	Sustainability (Switzerland), 17, 8 (2025)
29723	0.886	Faqir N.; Ennaji Y.; Chakir L.; Boumhidi J.	Hybrid Cnn-Lstm And Proximal Policy Optimization Model For Traffic Light Control In A Multi-Agent Environment	IEEE Access, 13 (2025)
40923	0.885	Zhang Z.; Zhou B.; Zhang B.; Cheng P.; Lee D.-H.; Hu S.	Optimizing Traffic Signal Control In Mixed Traffic Scenarios: A Predictive Traffic Information-Based Deep Reinforcement Learning Approach	2024 Forum for Innovative Sustainable Transportation Systems, FISTS 2024 (2024)
33264	0.883	Reda M.; Mountassir F.; Mohamed B.	Introduction To Coordinated Deep Agents For Traffic Signal	2019 International Conference on Wireless Technologies, Embedded and Intelligent Systems, WITS 2019 (2019)
23240	0.882	Dhanvijay M.M.; Patil S.C.	Energy Efficient Deep Reinforcement Learning Approach To Control The Traffic Flow In Iot Networks For Smart City	Journal of Ambient Intelligence and Humanized Computing, 15, 12 (2024)
50552	0.881	Lin T.; Lin R.	Smart City Traffic Flow And Signal Optimization Using Stgcn-Lstm And Ppo Algorithms	IEEE Access, 13 (2025)