中国空间科学技术

中国空间科学技术 ›› 2026, Vol. 46 ›› Issue (3): 232-243.doi: 10.16708/j.cnki.1000-758X.2026.0051

• 《中国空间科学技术（中英文）》创刊45周年专刊 • 上一篇    下一篇

基于遗传算法的空间碎片集持续观测布局优化

王钰淇，王鹏宇，郭延宁*   

  1. 哈尔滨工业大学航天学院，哈尔滨150001
  • 收稿日期:2025-12-17 修回日期:2026-04-09 录用日期:2026-04-13 发布日期:2026-05-21 出版日期:2026-05-31

Optimization of space debris set monitoring task layout based on genetic algorithm

WANG Yuqi，WANG Pengyu，Guo Yanning*   

  1. School of Astronautics,Harbin Institute of Technology,Harbin 150001,China
  • Received:2025-12-17 Revision received:2026-04-09 Accepted:2026-04-13 Online:2026-05-21 Published:2026-05-31

PDF (PC)

16

摘要： 针对卫星集群规模持续扩大、空间目标类型动态变化导致观测任务优化解空间维度呈指数级增长的问题，对空间碎片多角度、全方位持续观测技术进行研究，旨在提升卫星集群在观测时域内的观测覆盖效能、算法收敛稳定性及多场景适配性。构建了空间碎片观测区域数学表征模型，实现目标被观测区域与盲区的精准量化。设计了一种改进遗传算法，通过搭配自适应变异策略、轮盘赌选择算子、观测任务适应度函数，实现卫星集群初始位置部署与视场实时指向协同优化，以获取观测时域内的碎片表面有效特征信息。通过地球同步轨道上设计的碎片分散与碎片集中两组任务场景开展仿真验证。仿真结果表明，相较于蜣螂优化算法，基于遗传算法的空间碎片集监测任务布局优化算法收敛效果提高了10%，在不同卫星规模、目标密度场景下均达到82%以上的持续有效观测覆盖占比，具备高覆盖效率、强收敛稳定性及全场景适配能力。该算法可满足不同卫星规模下稀疏至密集空间碎片集的高效持续观测需求，为卫星集群空间碎片持续观测任务优化提供技术支撑。

关键词: 多星测控, 任务规划, 遗传算法, 高轨卫星, 测控调度

Abstract: To address the problem that the continuous expansion of satellite cluster scales and dynamic variations in space target types induce an exponential growth in the dimensionality of the optimal solution space for observation tasks, this study investigated multiangle and omnidirectional sustained observation technologies for space debris, with the aim of improving the observation coverage effectiveness, algorithm convergence stability and multi-scenario adaptability of satellite clusters within the observation time domain.A mathematical characterization model for the space debris observation region was constructed to realize the accurate quantification of the target observed area and blind zone.An improved genetic algorithm was subsequently designed by integrating an adaptive mutation strategy, a roulette-wheel selection operator and an observation task fitness function, thereby achieving the collaborative optimization of initial position deployment for satellite clusters and real-time field-of-view pointing to capture effective surface feature information of space debris throughout the observation time domain.Simulation verification was carried out in two task scenarios configured in the geostationary Earth orbit, namely dispersed debris and concentrated debris.The simulation results show that compared with the dung beetle optimization algorithm, the proposed layout optimization algorithm for space debris group monitoring tasks based on the improved genetic algorithm improves the convergence performance by 10%, achieves a sustained effective observation coverage ratio of more than 82% under scenarios with different satellite scales and target densities, and exhibits high coverage efficiency, strong convergence stability and full-scenario adaptability.This algorithm meets the efficient and sustained observation requirements of sparse to dense space debris groups with various satellite cluster scales, and provides technical support for the task optimization of sustained space debris observation using satellite clusters.

Key words: multi-satellite TT&, C；task planning；genetic algorithms；high-orbit satellite；TT&, C scheduling