Large language models improve agile satellite imaging schedules
Large Language Model-Assisted Adaptive Heuristic Search for Agile Earth Observation Satellite Scheduling
Artificial IntelligenceComputation and Language
Summary
Agile satellites have limited time and energy to take images of Earth, so they need smart plans to decide what to capture and when. This paper proposes a way to use large language models (LLMs) to create and improve rules for choosing imaging tasks during scheduling. The LLM suggests initial strategies, then learns from feedback during a search process to refine those strategies. The authors tested this method in different scenarios and found it led to better imaging schedules than previous methods.
What this means in practice
- •For satellite operations teams: Generate more efficient imaging schedules for agile Earth observation satellites under operational constraints.
- •For industrial robotics planners: Use adaptive heuristic improvements guided by language models to optimize task scheduling under resource limits.
Authors
He Wang, Yeye Liu, Ruijie Li, Xuemei Jiang, Junyu Wu, Lining Liu Zhuoqi Man, Tao Li, Yanjie Song
Abstract
Agile Earth observation satellites must coordinate imaging tasks within limited observation windows, while accounting for attitude maneuvering and onboard energy constraints. Competition among observation opportunities makes scheduling performance sensitive to the heuristic rules used for task selection. This letter proposes a large language model (LLM)-assisted adaptive heuristic search method for agile Earth observation satellite scheduling. The LLM generates an initial heuristic rule according to the characteristics of the scheduling instance. During the subsequent ant colony search, search performance and scheduling information are fed back to the LLM to refine the heuristic rule. The proposed modifications are evaluated through paired solution constructions before being incorporated into the search process. A unified scheduling procedure determines task selection, satellite assignment, and observation timing under the operational constraints. Experiments on six scheduling scenarios involving 50--200 imaging tasks demonstrate that the proposed method achieves 4.3\%--9.6\% higher mean observation benefit than the strongest baseline.