Task-Oriented Agentic AI-Driven Resource Orchestration in Satellite-Assisted IoV
IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 10559-10572
DOI: 10.1109/tccn.2026.3714048 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Ying Xie und 4 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Chongqing, Postgraduate Research Innovation Projects of Chongqing.
Abstract (Originalsprache)
Satellite-Assisted Internet of Vehicles (SIoV) has emerged as a promising paradigm to support heterogeneous vehicular applications with diverse quality of service (QoS) requirements by providing ubiquitous connectivity and computing. However, efficient resource orchestration in SIoV remains highly challenging due to the dynamic multi-layer topology and the uneven distribution of resources. To address these challenges, we propose a QoS-driven distributed resource orchestration scheme for SIoV. Specifically, we formulate a joint optimization problem that aims to maximize differentiated QoS for heterogeneous tasks, while incorporating a handover-aware offloading strategy to ensure service continuity under vehicular mobility. To efficiently solve the formulated problem, a task-oriented multi-head attention mechanism is integrated into multi-agent deep reinforcement learning to assess resource sufficiency across available nodes, enabling agents to select the most suitable execution node. Simulation results demonstrate that the proposed scheme significantly outperforms existing schemes in terms of QoS and exhibits superior scalability under varying vehicle densities.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Chongqing
- Postgraduate Research Innovation Projects of Chongqing
Zitieren
APA
Xie, Y., Song, Q., Sun, W., Guo, L., & Jamalipour, A. (2026). Task-Oriented Agentic AI-Driven Resource Orchestration in Satellite-Assisted IoV. IEEE Transactions on Cognitive Communications and Networking, 12, 10559-10572. https://doi.org/10.1109/tccn.2026.3714048
Vancouver
Xie Y, Song Q, Sun W, Guo L, Jamalipour A. Task-Oriented Agentic AI-Driven Resource Orchestration in Satellite-Assisted IoV. IEEE Transactions on Cognitive Communications and Networking. 2026;12:10559-10572. doi:10.1109/tccn.2026.3714048
BibTeX
@article{Xie2026,
author = {Xie, Ying and Song, Qingyang and Sun, Wei and Guo, Lei and Jamalipour, Abbas},
title = {Task-Oriented Agentic AI-Driven Resource Orchestration in Satellite-Assisted IoV},
journal = {IEEE Transactions on Cognitive Communications and Networking},
year = {2026},
volume = {12},
pages = {10559--10572},
doi = {10.1109/tccn.2026.3714048}
}
EUVIMED