Dueling Double Deep Q-Network for Adaptive Traffic Signal Control with Low Exhaust Emissions in A Single Intersection

Shu Fang; Feng Chen; Hongchao Liu

doi:10.1088/1757-899X/612/5/052039

Dueling Double Deep Q-Network for Adaptive Traffic Signal Control with Low Exhaust Emissions in A Single Intersection

Shu Fang, Feng Chen, Hongchao Liu

Civil Environ Construct Engineering

Research output: Contribution to journal › Conference article › peer-review

Abstract

In order to reduce traffic exhaust emissions caused by the large quantities of vehicles, this paper studied the traffic signal control (TSC) model with low exhaust emissions on the basis of the deep reinforcement learning. In this study, the Dueling Double DQN with prioritized replay (DDDQN-PR) algorithm we proposed was combined with the Double DQN, Dueling DQN, and prioritized replay to achieve the goal of low exhaust emissions of TSC. The agent was trained in traffic simulator USTCMTS2.1 in a single intersection. The experimental results show that the performance of DDDQN-PR was significantly better than the other four algorithms, not only in data efficiency but also in final performance.

Original language	English
Article number	052039
Journal	IOP Conference Series: Materials Science and Engineering
Volume	612
Issue number	5
DOIs	https://doi.org/10.1088/1757-899X/612/5/052039
State	Published - Oct 21 2019
Event	2019 6th International Conference on Advanced Composite Materials and Manufacturing Engineering, ACMME 2019 - Xishuangbanna, Yunnan, China Duration: Jun 22 2019 → Jun 23 2019

Access to Document

10.1088/1757-899X/612/5/052039

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Dueling Double Deep Q-Network for Adaptive Traffic Signal Control with Low Exhaust Emissions in A Single Intersection'. Together they form a unique fingerprint.

Cite this

@article{2f19d5011f91472dbc347896b5041afc,

title = "Dueling Double Deep Q-Network for Adaptive Traffic Signal Control with Low Exhaust Emissions in A Single Intersection",

abstract = "In order to reduce traffic exhaust emissions caused by the large quantities of vehicles, this paper studied the traffic signal control (TSC) model with low exhaust emissions on the basis of the deep reinforcement learning. In this study, the Dueling Double DQN with prioritized replay (DDDQN-PR) algorithm we proposed was combined with the Double DQN, Dueling DQN, and prioritized replay to achieve the goal of low exhaust emissions of TSC. The agent was trained in traffic simulator USTCMTS2.1 in a single intersection. The experimental results show that the performance of DDDQN-PR was significantly better than the other four algorithms, not only in data efficiency but also in final performance.",

author = "Shu Fang and Feng Chen and Hongchao Liu",

note = "Publisher Copyright: {\textcopyright} 2019 Published under licence by IOP Publishing Ltd. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; null ; Conference date: 22-06-2019 Through 23-06-2019",

year = "2019",

month = oct,

day = "21",

doi = "10.1088/1757-899X/612/5/052039",

language = "English",

volume = "612",

journal = "IOP Conference Series: Materials Science and Engineering",

issn = "1757-8981",

number = "5",

}

TY - JOUR

T1 - Dueling Double Deep Q-Network for Adaptive Traffic Signal Control with Low Exhaust Emissions in A Single Intersection

AU - Fang, Shu

AU - Chen, Feng

AU - Liu, Hongchao

PY - 2019/10/21

Y1 - 2019/10/21

N2 - In order to reduce traffic exhaust emissions caused by the large quantities of vehicles, this paper studied the traffic signal control (TSC) model with low exhaust emissions on the basis of the deep reinforcement learning. In this study, the Dueling Double DQN with prioritized replay (DDDQN-PR) algorithm we proposed was combined with the Double DQN, Dueling DQN, and prioritized replay to achieve the goal of low exhaust emissions of TSC. The agent was trained in traffic simulator USTCMTS2.1 in a single intersection. The experimental results show that the performance of DDDQN-PR was significantly better than the other four algorithms, not only in data efficiency but also in final performance.

AB - In order to reduce traffic exhaust emissions caused by the large quantities of vehicles, this paper studied the traffic signal control (TSC) model with low exhaust emissions on the basis of the deep reinforcement learning. In this study, the Dueling Double DQN with prioritized replay (DDDQN-PR) algorithm we proposed was combined with the Double DQN, Dueling DQN, and prioritized replay to achieve the goal of low exhaust emissions of TSC. The agent was trained in traffic simulator USTCMTS2.1 in a single intersection. The experimental results show that the performance of DDDQN-PR was significantly better than the other four algorithms, not only in data efficiency but also in final performance.

UR - http://www.scopus.com/inward/record.url?scp=85074404541&partnerID=8YFLogxK

U2 - 10.1088/1757-899X/612/5/052039

DO - 10.1088/1757-899X/612/5/052039

M3 - Conference article

AN - SCOPUS:85074404541

VL - 612

JO - IOP Conference Series: Materials Science and Engineering

JF - IOP Conference Series: Materials Science and Engineering

SN - 1757-8981

IS - 5

M1 - 052039

Y2 - 22 June 2019 through 23 June 2019

ER -