Research on Traffic Flow Forecasting Based on Dynamic Spatial-Temporal Transformer

IF 1.6 4区 工程技术 Q3 ENGINEERING, CIVIL
Hong Zhang, Hongyan Wang, Xijun Zhang, Lei Gong
{"title":"Research on Traffic Flow Forecasting Based on Dynamic Spatial-Temporal Transformer","authors":"Hong Zhang, Hongyan Wang, Xijun Zhang, Lei Gong","doi":"10.1177/03611981231205880","DOIUrl":null,"url":null,"abstract":"Accurate traffic flow forecasting is crucial for urban traffic control and route planning. Aiming at the difficulty in capturing dynamic spatio-temporal complexity of traffic flow, a dynamic spatio-temporal transformer (DST-Trans) model capable of modeling dynamic correlation of traffic flow is proposed, which consists of gated temporal convolutional network (GTCN), graph convolutional network (GCN), and spatio-temporal transformer (ST-TF). GTCN and GCN are utilized to capture the temporal and spatial characteristics of traffic flow, respectively. ST-TF includes a temporal transformer using temporal gated convolution and temporal multi-head self-attention to capture short-long term temporal features, and spatial transformer using spatial gated graph convolution and spatial multi-head self-attention to capture local-global dynamic spatial features. In addition, to take full advantage of the dynamic and static associations of road networks, multi-graph models of road relationship graph, similarity graph, and adaptive dynamic graph with SGGC are constructed. Experimental results show that the DST-Trans model in this paper shows good prediction performance in short-term (15 min), medium-term (30 min), and long-term (60 min) prediction, outperforming existing state-of-the-art models by up to approximately 7%.","PeriodicalId":23279,"journal":{"name":"Transportation Research Record","volume":" 7","pages":"0"},"PeriodicalIF":1.6000,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transportation Research Record","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1177/03611981231205880","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}
引用次数: 0

Abstract

Accurate traffic flow forecasting is crucial for urban traffic control and route planning. Aiming at the difficulty in capturing dynamic spatio-temporal complexity of traffic flow, a dynamic spatio-temporal transformer (DST-Trans) model capable of modeling dynamic correlation of traffic flow is proposed, which consists of gated temporal convolutional network (GTCN), graph convolutional network (GCN), and spatio-temporal transformer (ST-TF). GTCN and GCN are utilized to capture the temporal and spatial characteristics of traffic flow, respectively. ST-TF includes a temporal transformer using temporal gated convolution and temporal multi-head self-attention to capture short-long term temporal features, and spatial transformer using spatial gated graph convolution and spatial multi-head self-attention to capture local-global dynamic spatial features. In addition, to take full advantage of the dynamic and static associations of road networks, multi-graph models of road relationship graph, similarity graph, and adaptive dynamic graph with SGGC are constructed. Experimental results show that the DST-Trans model in this paper shows good prediction performance in short-term (15 min), medium-term (30 min), and long-term (60 min) prediction, outperforming existing state-of-the-art models by up to approximately 7%.
基于动态时空变换的交通流预测研究
准确的交通流预测对于城市交通控制和路线规划至关重要。针对交通流动态时空复杂性难以捕捉的问题,提出了一种能够对交通流动态相关性进行建模的动态时空转换器(DST-Trans)模型,该模型由门控时间卷积网络(GTCN)、图卷积网络(GCN)和时空转换器(ST-TF)组成。利用GTCN和GCN分别捕捉交通流的时空特征。ST-TF包括一个使用时间门控卷积和时间多头自注意捕获短期长期时间特征的时间转换器,以及使用空间门控图卷积和空间多头自注意捕获局部-全局动态空间特征的空间转换器。此外,为了充分利用路网的动态关联和静态关联,构建了基于SGGC的道路关系图、相似图和自适应动态图的多图模型。实验结果表明,本文的DST-Trans模型在短期(15分钟)、中期(30分钟)和长期(60分钟)预测中表现出良好的预测性能,比现有最先进的模型高出约7%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Transportation Research Record
Transportation Research Record 工程技术-工程:土木
CiteScore
3.20
自引率
11.80%
发文量
918
审稿时长
4.2 months
期刊介绍: Transportation Research Record: Journal of the Transportation Research Board is one of the most cited and prolific transportation journals in the world, offering unparalleled depth and breadth in the coverage of transportation-related topics. The TRR publishes approximately 70 issues annually of outstanding, peer-reviewed papers presenting research findings in policy, planning, administration, economics and financing, operations, construction, design, maintenance, safety, and more, for all modes of transportation. This site provides electronic access to a full compilation of papers since the 1996 series.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信