Mobile BS-Assisted Autonomous Scheduling of UAVs, Workers and Servers for Data Collection and Offloading in Disaster Response
IEEE Transactions on Mobile Computing · 2026-12 · 25 (12) · 22554-22571
DOI: 10.1109/tmc.2026.3723464 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2026-12. Verfasst von Yitong Guo und 8 weiteren Autor:innen. Die Arbeit nennt 73 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Postdoctoral Fellowship Program, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
Frequent natural disasters threaten public safety and economic development, while effective response relies on timely acquisition and processing of post-disaster environmental information, including task-related sensing data and on-site descriptions for situational assessment, rescue planning, and resource allocation. These processes face challenges from inefficient sensing, limited onboard resources, and damaged communication infrastructure. To address these challenges, we propose the Mobile Base Station-Assisted Heterogeneous Multi-Agent Online Autonomous Collaborative Scheduling Algorithm (MBAS-IGLN) to support efficient collection, offloading, and processing of post-disaster sensing data. The core concept of MBAS-IGLN lies in constructing a two-layer scheduling framework. (1) At the upper layer, mobile base stations perform dynamic game scheduling driven by information fields constructed through Gaussian diffusion, thereby providing stable and efficient communication and information assurance for lower-layer agents; (2) At the lower layer, UAVs, workers, and servers achieve autonomous collaborative scheduling within the communication coverage of mobile base stations through a local Nash equilibrium game. This enables matching for data collection and offloading while providing environmental feedback to the upper-layer mobile base stations. Finally, we conducted detailed experiments using extensive real-world and simulated data. Compared to baselines (GREEDY, K-WTA, MADL, and MARL), MBAS-IGLN improved data offloading rates by 40.16%, 31.46%, 19.30%, and 16.41%, respectively, while enabling online decisions within a few seconds in the small-scale benchmark scenarios, demonstrating its effectiveness in dynamic post-disaster environments.
Förderung
- National Natural Science Foundation of China
- Postdoctoral Fellowship Program
- China Postdoctoral Science Foundation
- Natural Science Basic Research Program of Shaanxi
- Fundamental Research Funds for the Central Universities
- Shaanxi Key Technology R D Program
- Natural Science Funds for Distinguished Young Scholar of Shaanxi
Zitieren
APA
Guo, Y., Han, L., Zhao, M., Li, Q., Wang, L., Wan, B., Wang, Q., Guo, B., & Yang, P. (2026). Mobile BS-Assisted Autonomous Scheduling of UAVs, Workers and Servers for Data Collection and Offloading in Disaster Response. IEEE Transactions on Mobile Computing, 25(12), 22554-22571. https://doi.org/10.1109/tmc.2026.3723464
Vancouver
Guo Y, Han L, Zhao M, Li Q, Wang L, Wan B, et al. Mobile BS-Assisted Autonomous Scheduling of UAVs, Workers and Servers for Data Collection and Offloading in Disaster Response. IEEE Transactions on Mobile Computing. 2026;25(12):22554-22571. doi:10.1109/tmc.2026.3723464
BibTeX
@article{Guo2026,
author = {Guo, Yitong and Han, Lei and Zhao, Mengting and Li, Qingyang and Wang, Liang and Wan, Bo and Wang, Quan and Guo, Bin and Yang, Pengfei},
title = {Mobile BS-Assisted Autonomous Scheduling of UAVs, Workers and Servers for Data Collection and Offloading in Disaster Response},
journal = {IEEE Transactions on Mobile Computing},
year = {2026},
volume = {25},
number = {12},
pages = {22554--22571},
doi = {10.1109/tmc.2026.3723464}
}
EUVIMED