LLM Agent-Enhanced Bi-Level Online Order Dispatching and Scheduling for Large-Scale Agile Earth Observation Satellites
IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 1001916-1001916
DOI: 10.1109/tgrs.2026.3710264 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Ziyang Zhang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Innovation Program of Hunan Province, Young Elite Scientists Sponsorship Program by China Academy of Space Technology.
Abstract (Originalsprache)
With the growth of remote sensing satellites, large-scale agile Earth observation satellite online order dispatching and scheduling (LAEOS-OODS) has emerged as a new type of resource optimization problem, which is characterized by super-large scale and real-time requirements in the aerospace domain. To address this problem, this article proposes an LLM agent-enhanced bi-level online order dispatching and scheduling (LLM-OODS) algorithm that consists of four agents: a dispatcher in chief (DIC), a single-satellite scheduler, a rescheduler, and an evaluator. Within the bi-level algorithm framework, the DIC is enhanced by an LLM in the upper layer and dispatches online orders to satellites. In the lower layer, the single-satellite scheduler performs order optimization and feedback using an adaptive large neighborhood search (ALNS) metaheuristic improved by a backward-relax strategy. To address the unscheduled orders, the rescheduler executes heuristic satellite traversal and efficient order insertion. Working with the aforementioned agents, the evaluator conducts analysis, feedback, and improvement. Experiment results indicate that the proposed algorithm outperforms other state-of-the-art algorithms and traditional methods in varying-scale LAEOS-OODS; also, the agents in the algorithm show respective outperformance. This article presents good potential of LLM enhancement for large-scale real-world resource optimization.
Förderung
- National Natural Science Foundation of China
- Science and Technology Innovation Program of Hunan Province
- Young Elite Scientists Sponsorship Program by China Academy of Space Technology
Zitieren
APA
Zhang, Z., Du, Y., Song, Y., Liu, X., Chen, J., Zhu, Z., & Xing, L. (2026). LLM Agent-Enhanced Bi-Level Online Order Dispatching and Scheduling for Large-Scale Agile Earth Observation Satellites. IEEE Transactions on Geoscience and Remote Sensing, 64, 1001916-1001916. https://doi.org/10.1109/tgrs.2026.3710264
Vancouver
Zhang Z, Du Y, Song Y, Liu X, Chen J, Zhu Z, et al. LLM Agent-Enhanced Bi-Level Online Order Dispatching and Scheduling for Large-Scale Agile Earth Observation Satellites. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:1001916-1001916. doi:10.1109/tgrs.2026.3710264
BibTeX
@article{Zhang2026,
author = {Zhang, Ziyang and Du, Yonghao and Song, Yanjie and Liu, Xiaolu and Chen, Jingfang and Zhu, Zhengqiu and Xing, Lining},
title = {LLM Agent-Enhanced Bi-Level Online Order Dispatching and Scheduling for Large-Scale Agile Earth Observation Satellites},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
year = {2026},
volume = {64},
pages = {1001916--1001916},
doi = {10.1109/tgrs.2026.3710264}
}
EUVIMED