Environmental Adaptive Urban Traffic Signal Control Based on Reinforcement Learning Algorithm

Urban traffic signal control is an important part of the construction of intelligent regional traffic. Aiming at the problem of the optimal control strategy in urban traffic signals, this paper proposes environmental adaptive urban traffic signal control based on reinforcement learning algorithms. T...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series Vol. 1650; no. 3; pp. 32097 - 32102
Main Authors: Feng, Yanming, Wu, Yongrong
Format: Journal Article
Language:English
Published: Bristol IOP Publishing 01.10.2020
Subjects:
Adaptive control
Algorithms
Deep learning
environment adaptive
Machine learning
Optimal control
Optimization
Physics
Queues
reinforcement learning
Traffic control
Traffic engineering
Traffic flow
Traffic signal control
Traffic signals
Traffic speed
ISSN:1742-6588, 1742-6596
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Urban traffic signal control is an important part of the construction of intelligent regional traffic. Aiming at the problem of the optimal control strategy in urban traffic signals, this paper proposes environmental adaptive urban traffic signal control based on reinforcement learning algorithms. Through the continuous perception of the traffic environment, the position and speed of the vehicles in different environments are expressed in a matrix, and the parameters are continuously iteratively optimized through the reinforcement learning method to optimize the objective function (the vehicle that can pass the most in a limited time) to achieve the purpose of effective vehicle control. According to the traffic simulation software Vissim, it can be known that the algorithm proposed in this paper performs better in terms of average waiting queue length and global average speed compared with other algorithms, and the deep learning algorithm is significantly better than other algorithms in terms of stability. The average speed of the deep learning algorithm is increased by 9% compared with the baseline, and the average waiting queue length is reduced by 13.4% compared with the baseline. The experiments studied this time are sufficient to prove that the algorithm in this paper can adapt to the dynamically changing and complex urban traffic environment and has great research value.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1650/3/032097