Deep Reinforcement Learning-based Traffic Signal Control

Junyun Ruan; Jinzhuo Tang; Ge Gao; Tianyu Shi; Alaa Khamis

doi:10.1109/SM57895.2023.10112300

Deep Reinforcement Learning-based Traffic Signal Control

Junyun Ruan
, Jinzhuo Tang
, Ge Gao
, Tianyu Shi
, Alaa Khamis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

Traffic congestion has become an increasingly concerning problem in modern society. Recent research has proven that Reinforcement Learning (RL) applied to Traffic Signal Control (TSC) is useful in mitigating congestion. In this paper, a model of real-world intersection with real traffic data collected in Hangzhou, China is simulated with different RL based traffic signal controllers. Two model free reinforcement learning methods are proposed namely: Deep Q-Learning (DQN) and double DQN (DDQN). These models are trained and tested at a 4-way intersection. Model adaptability and performance in different traffic scenarios are also measured and discussed in this paper.

Original language	English
Title of host publication	2023 IEEE International Conference on Smart Mobility, SM 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	21-26
Number of pages	6
ISBN (Electronic)	9798350312751
DOIs	https://doi.org/10.1109/SM57895.2023.10112300
State	Published - 2023
Externally published	Yes
Event	2023 IEEE International Conference on Smart Mobility, SM 2023 - Thuwal, Saudi Arabia Duration: 19 Mar 2023 → 21 Mar 2023

Publication series

Name	2023 IEEE International Conference on Smart Mobility, SM 2023

Conference

Conference	2023 IEEE International Conference on Smart Mobility, SM 2023
Country/Territory	Saudi Arabia
City	Thuwal
Period	19/03/23 → 21/03/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

Deep Learning
Reinforcement Leaning
Traffic Congestion
Traffic Signal Control

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Vision and Pattern Recognition
Control and Optimization
Transportation

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/SM57895.2023.10112300

Cite this

@inproceedings{1bde2201c38a4ed0a3cbd3c8ee821217,

title = "Deep Reinforcement Learning-based Traffic Signal Control",

abstract = "Traffic congestion has become an increasingly concerning problem in modern society. Recent research has proven that Reinforcement Learning (RL) applied to Traffic Signal Control (TSC) is useful in mitigating congestion. In this paper, a model of real-world intersection with real traffic data collected in Hangzhou, China is simulated with different RL based traffic signal controllers. Two model free reinforcement learning methods are proposed namely: Deep Q-Learning (DQN) and double DQN (DDQN). These models are trained and tested at a 4-way intersection. Model adaptability and performance in different traffic scenarios are also measured and discussed in this paper.",

keywords = "Deep Learning, Reinforcement Leaning, Traffic Congestion, Traffic Signal Control",

author = "Junyun Ruan and Jinzhuo Tang and Ge Gao and Tianyu Shi and Alaa Khamis",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Conference on Smart Mobility, SM 2023 ; Conference date: 19-03-2023 Through 21-03-2023",

year = "2023",

doi = "10.1109/SM57895.2023.10112300",

language = "English",

series = "2023 IEEE International Conference on Smart Mobility, SM 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "21--26",

booktitle = "2023 IEEE International Conference on Smart Mobility, SM 2023",

address = "United States",

}

Ruan, J, Tang, J, Gao, G, Shi, T & Khamis, A 2023, Deep Reinforcement Learning-based Traffic Signal Control. in 2023 IEEE International Conference on Smart Mobility, SM 2023. 2023 IEEE International Conference on Smart Mobility, SM 2023, Institute of Electrical and Electronics Engineers Inc., pp. 21-26, 2023 IEEE International Conference on Smart Mobility, SM 2023, Thuwal, Saudi Arabia, 19/03/23. https://doi.org/10.1109/SM57895.2023.10112300

Deep Reinforcement Learning-based Traffic Signal Control. / Ruan, Junyun; Tang, Jinzhuo; Gao, Ge et al.
2023 IEEE International Conference on Smart Mobility, SM 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 21-26 (2023 IEEE International Conference on Smart Mobility, SM 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Deep Reinforcement Learning-based Traffic Signal Control

AU - Ruan, Junyun

AU - Tang, Jinzhuo

AU - Gao, Ge

AU - Shi, Tianyu

AU - Khamis, Alaa

PY - 2023

Y1 - 2023

N2 - Traffic congestion has become an increasingly concerning problem in modern society. Recent research has proven that Reinforcement Learning (RL) applied to Traffic Signal Control (TSC) is useful in mitigating congestion. In this paper, a model of real-world intersection with real traffic data collected in Hangzhou, China is simulated with different RL based traffic signal controllers. Two model free reinforcement learning methods are proposed namely: Deep Q-Learning (DQN) and double DQN (DDQN). These models are trained and tested at a 4-way intersection. Model adaptability and performance in different traffic scenarios are also measured and discussed in this paper.

AB - Traffic congestion has become an increasingly concerning problem in modern society. Recent research has proven that Reinforcement Learning (RL) applied to Traffic Signal Control (TSC) is useful in mitigating congestion. In this paper, a model of real-world intersection with real traffic data collected in Hangzhou, China is simulated with different RL based traffic signal controllers. Two model free reinforcement learning methods are proposed namely: Deep Q-Learning (DQN) and double DQN (DDQN). These models are trained and tested at a 4-way intersection. Model adaptability and performance in different traffic scenarios are also measured and discussed in this paper.

KW - Deep Learning

KW - Reinforcement Leaning

KW - Traffic Congestion

KW - Traffic Signal Control

UR - https://www.scopus.com/pages/publications/85159857165

U2 - 10.1109/SM57895.2023.10112300

DO - 10.1109/SM57895.2023.10112300

M3 - Conference contribution

AN - SCOPUS:85159857165

T3 - 2023 IEEE International Conference on Smart Mobility, SM 2023

SP - 21

EP - 26

BT - 2023 IEEE International Conference on Smart Mobility, SM 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE International Conference on Smart Mobility, SM 2023

Y2 - 19 March 2023 through 21 March 2023

ER -

Deep Reinforcement Learning-based Traffic Signal Control

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Fingerprint

Cite this