Presslight: Learning Max pressure control to coordinate traffic signals in arterial network

Hua Wei, Chacha Chen, Guanjie Zheng, Kan Wu, Vikash Gayah, Kai Xu, Zhenhui Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Traffic signal control is essential for transportation efficiency in road networks. It has been a challenging problem because of the complexity in traffic dynamics. Conventional transportation research suffers from the incompetency to adapt to dynamic traffic situations. Recent studies propose to use reinforcement learning (RL) to search for more efficient traffic signal plans. However, most existing RL-based studies design the key elements - reward and state - in a heuristic way. This results in highly sensitive performances and a long learning process. To avoid the heuristic design of RL elements, we propose to connect RL with recent studies in transportation research. Our method is inspired by the state-of-the-art method max pressure (MP) in the transportation field. The reward design of our method is well supported by the theory in MP, which can be proved to be maximizing the throughput of the traffic network, i.e., minimizing the overall network travel time. We also show that our concise state representation can fully support the optimization of the proposed reward function. Through comprehensive experiments, we demonstrate that our method outperforms both conventional transportation approaches and existing learning-based methods.

Original languageEnglish (US)
Title of host publicationKDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages1290-1298
Number of pages9
ISBN (Electronic)9781450362016
DOIs
StatePublished - Jul 25 2019
Event25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019 - Anchorage, United States
Duration: Aug 4 2019Aug 8 2019

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Conference

Conference25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019
CountryUnited States
CityAnchorage
Period8/4/198/8/19

Fingerprint

Traffic signals
Pressure control
Reinforcement learning
Travel time
Throughput
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems

Cite this

Wei, H., Chen, C., Zheng, G., Wu, K., Gayah, V., Xu, K., & Li, Z. (2019). Presslight: Learning Max pressure control to coordinate traffic signals in arterial network. In KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1290-1298). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3292500.3330949
Wei, Hua ; Chen, Chacha ; Zheng, Guanjie ; Wu, Kan ; Gayah, Vikash ; Xu, Kai ; Li, Zhenhui. / Presslight : Learning Max pressure control to coordinate traffic signals in arterial network. KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2019. pp. 1290-1298 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
@inproceedings{94d0396a60e14877b707de6808ec9af8,
title = "Presslight: Learning Max pressure control to coordinate traffic signals in arterial network",
abstract = "Traffic signal control is essential for transportation efficiency in road networks. It has been a challenging problem because of the complexity in traffic dynamics. Conventional transportation research suffers from the incompetency to adapt to dynamic traffic situations. Recent studies propose to use reinforcement learning (RL) to search for more efficient traffic signal plans. However, most existing RL-based studies design the key elements - reward and state - in a heuristic way. This results in highly sensitive performances and a long learning process. To avoid the heuristic design of RL elements, we propose to connect RL with recent studies in transportation research. Our method is inspired by the state-of-the-art method max pressure (MP) in the transportation field. The reward design of our method is well supported by the theory in MP, which can be proved to be maximizing the throughput of the traffic network, i.e., minimizing the overall network travel time. We also show that our concise state representation can fully support the optimization of the proposed reward function. Through comprehensive experiments, we demonstrate that our method outperforms both conventional transportation approaches and existing learning-based methods.",
author = "Hua Wei and Chacha Chen and Guanjie Zheng and Kan Wu and Vikash Gayah and Kai Xu and Zhenhui Li",
year = "2019",
month = "7",
day = "25",
doi = "10.1145/3292500.3330949",
language = "English (US)",
series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",
pages = "1290--1298",
booktitle = "KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

Wei, H, Chen, C, Zheng, G, Wu, K, Gayah, V, Xu, K & Li, Z 2019, Presslight: Learning Max pressure control to coordinate traffic signals in arterial network. in KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 1290-1298, 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019, Anchorage, United States, 8/4/19. https://doi.org/10.1145/3292500.3330949

Presslight : Learning Max pressure control to coordinate traffic signals in arterial network. / Wei, Hua; Chen, Chacha; Zheng, Guanjie; Wu, Kan; Gayah, Vikash; Xu, Kai; Li, Zhenhui.

KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2019. p. 1290-1298 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Presslight

T2 - Learning Max pressure control to coordinate traffic signals in arterial network

AU - Wei, Hua

AU - Chen, Chacha

AU - Zheng, Guanjie

AU - Wu, Kan

AU - Gayah, Vikash

AU - Xu, Kai

AU - Li, Zhenhui

PY - 2019/7/25

Y1 - 2019/7/25

N2 - Traffic signal control is essential for transportation efficiency in road networks. It has been a challenging problem because of the complexity in traffic dynamics. Conventional transportation research suffers from the incompetency to adapt to dynamic traffic situations. Recent studies propose to use reinforcement learning (RL) to search for more efficient traffic signal plans. However, most existing RL-based studies design the key elements - reward and state - in a heuristic way. This results in highly sensitive performances and a long learning process. To avoid the heuristic design of RL elements, we propose to connect RL with recent studies in transportation research. Our method is inspired by the state-of-the-art method max pressure (MP) in the transportation field. The reward design of our method is well supported by the theory in MP, which can be proved to be maximizing the throughput of the traffic network, i.e., minimizing the overall network travel time. We also show that our concise state representation can fully support the optimization of the proposed reward function. Through comprehensive experiments, we demonstrate that our method outperforms both conventional transportation approaches and existing learning-based methods.

AB - Traffic signal control is essential for transportation efficiency in road networks. It has been a challenging problem because of the complexity in traffic dynamics. Conventional transportation research suffers from the incompetency to adapt to dynamic traffic situations. Recent studies propose to use reinforcement learning (RL) to search for more efficient traffic signal plans. However, most existing RL-based studies design the key elements - reward and state - in a heuristic way. This results in highly sensitive performances and a long learning process. To avoid the heuristic design of RL elements, we propose to connect RL with recent studies in transportation research. Our method is inspired by the state-of-the-art method max pressure (MP) in the transportation field. The reward design of our method is well supported by the theory in MP, which can be proved to be maximizing the throughput of the traffic network, i.e., minimizing the overall network travel time. We also show that our concise state representation can fully support the optimization of the proposed reward function. Through comprehensive experiments, we demonstrate that our method outperforms both conventional transportation approaches and existing learning-based methods.

UR - http://www.scopus.com/inward/record.url?scp=85071183003&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071183003&partnerID=8YFLogxK

U2 - 10.1145/3292500.3330949

DO - 10.1145/3292500.3330949

M3 - Conference contribution

AN - SCOPUS:85071183003

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 1290

EP - 1298

BT - KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -

Wei H, Chen C, Zheng G, Wu K, Gayah V, Xu K et al. Presslight: Learning Max pressure control to coordinate traffic signals in arterial network. In KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2019. p. 1290-1298. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). https://doi.org/10.1145/3292500.3330949