i-CLTP: Integrated contrastive learning with transformer framework for traffic state prediction and network-wide analysis
Artikel i vetenskaplig tidskrift, 2025

Traffic state predictions are critical for the traffic management and control of transport systems. This study introduces an innovative contrastive learning framework coupled with a transformer architecture for spatiotemporal traffic state prediction, designed to capture the spatio-temporal heterogeneity inherent in traffic. The transformer structure functions as the upper level of the prediction framework to minimize the prediction errors between the input and predicted output. Based on the self-supervised contrastive learning, the lower level in the framework is proposed to discern the spatio-temporal heterogeneity and embed the latent characteristic of traffic flow by regenerating the augmentation features. Then, a soft clustering problem is applied between the upper level and lower level to category the types of traffic flow characteristics by minimizing the joint loss across each cluster. Subsequently, the proposed model is evaluated through a real-world highway traffic flow dataset for bench marking against several latest existing models. The experimental results affirm that the proposed model considerably enhances traffic state prediction accuracy. In terms of precision metrics, the model records a Mean Absolute Error of 13.31 and a Mean Absolute Percentage Error of 7.85%, reflecting marked improvements of 2.0% and 14.5% respectively over the latest and most competitive baseline model. Furthermore, the analysis reveals that capacity of the proposed method to learn the cluster patterns of spatio-temporal traffic dynamics reflected by calibrated fundamental diagrams.

Soft clustering

Traffic state prediction

Fundamental diagram

Transformer

Contrastive learning

Författare

Ruo Jia

Chalmers, Arkitektur och samhällsbyggnadsteknik, Geologi och geoteknik

Kun Gao

Chalmers, Arkitektur och samhällsbyggnadsteknik, Geologi och geoteknik

Yang Liu

Chalmers, Arkitektur och samhällsbyggnadsteknik, Geologi och geoteknik

Bo Yu

Tongji University

Xiaolei Ma

Beihang University

Zhenliang Ma

Kungliga Tekniska Högskolan (KTH)

Transportation Research, Part C: Emerging Technologies

0968-090X (ISSN)

Vol. 171 104979

Digitala verktyg för hållbar planering och hantering av delad mikromobilitet med hjälp av Big Data

VINNOVA (2023-01042), 2023-09-04 -- 2025-03-31.

Eldrivna multimodala transportsystem för att stärka urban tillgänglighet och konnektivitet (eMATS)

Europeiska kommissionen (EU), 2023-01-01 -- 2025-12-31.

Energimyndigheten (2023-00029), 2023-05-05 -- 2026-04-30.

Styrkeområden

Transport

Ämneskategorier (SSIF 2011)

Transportteknik och logistik

Datavetenskap (datalogi)

DOI

10.1016/j.trc.2024.104979

Mer information

Senast uppdaterat

2025-01-10