Understanding Traffic Congestion via Traffic Attacker: A Traffic Demand and Supply View
Artificial Intelligence for Engineering · 2025
DOI: 10.1049/aie2.12001 ↗Kurz gefasst
Fachartikel in der Zeitschrift Artificial Intelligence for Engineering, 2025. Verfasst von Guanjie Zheng und 3 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Traffic congestion is an escalating public concern. Where is the bottleneck under congestion in an urban traffic system? To answer this question, we consider the traffic system as a demand supply system with the volume of vehicles on the road as the demand and the capacity of the road network as the supply. Then, we formulate an adversarial attack task in the traffic system which aims to perform traffic attacks that can induce greater congestion to test which parts are more vulnerable to congestion. To solve this problem, we propose T raffic A ttacker via R eceptive A ttention (TARA), an attention‐based model that captures complex patterns in the traffic system, to generate powerful traffic attacks. TARA enjoys high expressiveness to model dynamic correlations between different parts of the traffic system and great scalability on large road networks. Extensive experiments are conducted and TARA outperforms baseline methods in the proposed task. In addition, case studies illustrate the soundness of the demand and supply view and explain why the attack induced by TARA can cause more congestion.
Zitieren
APA
Zheng, G., Zhang, Y., Liu, Z., & Li, Z. (2025). Understanding Traffic Congestion via Traffic Attacker: A Traffic Demand and Supply View. Artificial Intelligence for Engineering. https://doi.org/10.1049/aie2.12001
Vancouver
Zheng G, Zhang Y, Liu Z, Li Z. Understanding Traffic Congestion via Traffic Attacker: A Traffic Demand and Supply View. Artificial Intelligence for Engineering. 2025. doi:10.1049/aie2.12001
BibTeX
@article{Zheng2025,
author = {Zheng, Guanjie and Zhang, Yuxing and Liu, Zhanyu and Li, Zhenhui},
title = {Understanding Traffic Congestion via Traffic Attacker: A Traffic Demand and Supply View},
journal = {Artificial Intelligence for Engineering},
year = {2025},
doi = {10.1049/aie2.12001}
}
EUVIMED