Adaptive Traffic Light Control With Deep Reinforcement Learning: An Evaluation of Traffic Flow and Energy Consumption

被引:4
|
作者
Koch, Lucas [1 ]
Brinkmann, Tobias [1 ]
Wegener, Marius [1 ]
Badalian, Kevin [1 ]
Andert, Jakob [1 ]
机构
[1] Rhein Westfal TH Aachen, Teaching & Res Area Mechatron Mobile Prop, D-52074 Aachen, Germany
关键词
Adaptive traffic light control; reinforcement learning; powertrain modeling; microscopic traffic simulation; GREEN-WAVE; SIGNAL CONTROL;
D O I
10.1109/TITS.2023.3305548
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Cities of all sizes around the world are facing increasing levels of congestion, which leads to increasing travel time and emissions, and ultimately affects the quality of life. Relevant research suggests that adaptive traffic light control systems can improve the traffic flow, but their impact on energy-efficiency of vehicle propulsion systems is not well understood. In this study, we use Proximal Policy Optimization, a Deep Reinforcement Learning algorithm, to develop an optimized adaptive traffic light control systems that controls three traffic lights simultaneously. For this purpose, we have created a microscopic traffic simulation of the city of Aachen, Germany, calibrated on the basis of traffic measurements, where the actual traffic light schedule, a green-wave, fixed-time control scheme, serves as a reference. The traffic simulation is coupled with detailed, physics-based powertrain models of both conventional and electric vehicles, which are validated against chassis dynamometer measurements. By analyzing the complex interactions between traffic light control, the resulting vehicle trajectories and the powertrain components, we show that Reinforcement Learning-based adaptive control can significantly improve the traffic flow, with a 41% increase in average velocity, without any drawbacks in CO2 emission (-1%). Furthermore, we find that maximizing traffic flow and minimizing CO2 emissions are not necessarily contradictory objectives, and identify an increased energy saving potential at low traffic densities. Thus, we prove that adaptive traffic light control can make traffic not only more time-efficient, but also more sustainable.
引用
收藏
页码:15066 / 15076
页数:11
相关论文
共 50 条
  • [21] Adaptive Traffic Signal Control Model on Intersections Based on Deep Reinforcement Learning
    Li, Duowei
    Wu, Jianping
    Xu, Ming
    Wang, Ziheng
    Hu, Kezhen
    [J]. Journal of Advanced Transportation, 2020, 2020
  • [22] Adaptive Traffic Signal Control Model on Intersections Based on Deep Reinforcement Learning
    Li, Duowei
    Wu, Jianping
    Xu, Ming
    Wang, Ziheng
    Hu, Kezhen
    [J]. JOURNAL OF ADVANCED TRANSPORTATION, 2020, 2020
  • [23] Deep Reinforcement Learning for Traffic Light Timing Optimization
    Wang, Bin
    He, Zhengkun
    Sheng, Jinfang
    Chen, Yu
    [J]. PROCESSES, 2022, 10 (11)
  • [24] Multimodal Traffic Light Control with Connected Vehicles: A Deep Reinforcement Learning Approach
    Zhou, Runhao
    Nousch, Tobias
    Adam, Django
    Hirrle, Angelika
    Wang, Meng
    [J]. 2023 8TH INTERNATIONAL CONFERENCE ON MODELS AND TECHNOLOGIES FOR INTELLIGENT TRANSPORTATION SYSTEMS, MT-ITS, 2023,
  • [25] Learning to Communicate with Reinforcement Learning for an Adaptive Traffic Control System
    Vanneste, Simon
    de Borrekens, Gauthier
    Bosmans, Stig
    Vanneste, Astrid
    Mets, Kevin
    Mercelis, Siegfried
    Latre, Steven
    Hellinckx, Peter
    [J]. ADVANCES ON P2P, PARALLEL, GRID, CLOUD AND INTERNET COMPUTING, 3PGCIC-2021, 2022, 343 : 207 - 216
  • [26] Reinforcement learning for True Adaptive traffic signal control
    Abdulhai, B
    Pringle, R
    Karakoulas, GJ
    [J]. JOURNAL OF TRANSPORTATION ENGINEERING, 2003, 129 (03) : 278 - 285
  • [27] A Deep Reinforcement Learning Approach to Traffic Signal Control
    Razack, Aquib Junaid
    Ajith, Vysyakh
    Gupta, Rajiv
    [J]. 2021 IEEE CONFERENCE ON TECHNOLOGIES FOR SUSTAINABILITY (SUSTECH2021), 2021,
  • [28] Deep Reinforcement Learning for Traffic Signal Control: A Review
    Rasheed, Faizan
    Yau, Kok-Lim Alvin
    Noor, Rafidah Md.
    Wu, Celimuge
    Low, Yeh-Ching
    [J]. IEEE ACCESS, 2020, 8 : 208016 - 208044
  • [29] A Survey on Deep Reinforcement Learning for Traffic Signal Control
    Miao, Wei
    Li, Long
    Wang, Zhiwen
    [J]. PROCEEDINGS OF THE 33RD CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2021), 2021, : 1092 - 1097
  • [30] Robust Deep Reinforcement Learning for Traffic Signal Control
    Kai Liang Tan
    Anuj Sharma
    Soumik Sarkar
    [J]. Journal of Big Data Analytics in Transportation, 2020, 2 (3): : 263 - 274