EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Route-Assisted Peer Model Training among Connected Vehicles under Different Radio Capabilities

Han Zheng, Mengjing Liu, Fan Ye, Yuanyuan Yang

ACM Transactions on Internet of Things · 30.11.2026 · 7 (4) · 1-31

DOI: 10.1145/3803413 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Internet of Things, online erschienen am 22.05.2026. Verfasst von Han Zheng und 3 weiteren Autor:innen. Die Arbeit nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: US National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fully decentralized model training for on-road vehicles enables leveraging crowdsourced data without relying on central servers, infrastructure, or persistent Internet connectivity. However, real-world vehicular scenarios pose fundamental challenges to decentralized learning, including highly dynamic network topology, unreliable wireless links, and heterogeneous radio capability constraints. To address these challenges, we propose RoADTrain , a route-assisted decentralized peer model training framework with formal convergence guarantees. In RoADTrain, vehicles share short-horizon route information to estimate inter-vehicle contact duration and link reliability, which are embedded into a base graph. Vehicles then select communication partners by maximizing the algebraic connectivity of the resulting communication graph, thereby accelerating information mixing and model convergence. We further investigate subcarrier reuse and extend the algorithm to three representative radio capability cases. Extensive evaluations show that RoADTrain achieves convergence comparable to communication-unconstrained state-of-the-art approaches while reducing communication overhead by up to 4.5×, and significantly outperforms communication-constrained decentralized baselines by up to 29% higher driving success rates in online evaluations. We further empirically demonstrate the robustness of RoADTrain under a range of unexpected effects encountered in complex real-world applications, as well as in a large-scale traffic scenario.

Förderung

  • US National Science Foundation

Zitieren

APA

Zheng, H., Liu, M., Ye, F., & Yang, Y. (2026). Route-Assisted Peer Model Training among Connected Vehicles under Different Radio Capabilities. ACM Transactions on Internet of Things, 7(4), 1-31. https://doi.org/10.1145/3803413

Vancouver

Zheng H, Liu M, Ye F, Yang Y. Route-Assisted Peer Model Training among Connected Vehicles under Different Radio Capabilities. ACM Transactions on Internet of Things. 2026;7(4):1-31. doi:10.1145/3803413

BibTeX
@article{Zheng2026,
  author = {Zheng, Han and Liu, Mengjing and Ye, Fan and Yang, Yuanyuan},
  title = {Route-Assisted Peer Model Training among Connected Vehicles under Different Radio Capabilities},
  journal = {ACM Transactions on Internet of Things},
  year = {2026},
  volume = {7},
  number = {4},
  pages = {1--31},
  doi = {10.1145/3803413}
}
Kostenlos starten →