EUVIMEDEUVIMED entdecken
Crossref

Attention-Based Deep Reinforcement Learning for Logistics and Data Collection in AoI-Aware UAV Networks

Zhongliang Li, Zhihao Li, Gaojie Chen, Chong Huang, Zheng Yang, Zhu Han

IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 9656-9670

DOI: 10.1109/tccn.2026.3707190 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Zhongliang Li und 5 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Gefördert von: Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China, Fundamental Research Funds for the Central Universities, Sun Yat-sen University, Natural Science Foundation of Fujian Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid development of the low-altitude economy (LAE), the application of uncrewed aerial vehicles (UAVs) in data collection and logistics delivery is witnessing significant expansion. This paper proposes an innovative UAV-enabled system, where UAVs are assumed to perform simultaneous data collection and last-mile delivery tasks. The system is required to address two critical optimization problems: maximizing the communication rate and minimizing the maximum delivery time, while ensuring data freshness measured by the age of information (AoI) and maintaining the fairness in non-orthogonal multiple access (NOMA). The optimization problems are non-convex and involve conflicting objectives. Given these complexities, we propose an attention-based multi-agent policy optimization (A-MAPPO) method to address the trajectory planning problem. In addition, to address the issue of high-dimensional input state space, we explored the application of attention mechanisms and hierarchical reinforcement learning (HRL) to facilitate the exploration and learning process. Simulation results demonstrate that the proposed A-MAPPO method outperforms existing algorithms such as MAPPO and PPO in terms of convergence speed, delivery time, and total communication while ensuring user fairness and AoI requirements.

Förderung

  • Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China
  • Fundamental Research Funds for the Central Universities, Sun Yat-sen University
  • Natural Science Foundation of Fujian Province
  • National Science Foundation
  • Japan Science and Technology Agency (JST) Adopting Sustainable Partnerships for Innovative Research Ecosystem

Zitieren

APA

Li, Z., Li, Z., Chen, G., Huang, C., Yang, Z., & Han, Z. (2026). Attention-Based Deep Reinforcement Learning for Logistics and Data Collection in AoI-Aware UAV Networks. IEEE Transactions on Cognitive Communications and Networking, 12, 9656-9670. https://doi.org/10.1109/tccn.2026.3707190

Vancouver

Li Z, Li Z, Chen G, Huang C, Yang Z, Han Z. Attention-Based Deep Reinforcement Learning for Logistics and Data Collection in AoI-Aware UAV Networks. IEEE Transactions on Cognitive Communications and Networking. 2026;12:9656-9670. doi:10.1109/tccn.2026.3707190

BibTeX
@article{Li2026,
  author = {Li, Zhongliang and Li, Zhihao and Chen, Gaojie and Huang, Chong and Yang, Zheng and Han, Zhu},
  title = {Attention-Based Deep Reinforcement Learning for Logistics and Data Collection in AoI-Aware UAV Networks},
  journal = {IEEE Transactions on Cognitive Communications and Networking},
  year = {2026},
  volume = {12},
  pages = {9656--9670},
  doi = {10.1109/tccn.2026.3707190}
}
Kostenlos starten →