航空学报 > 2026, Vol. 47 Issue (14): 332728-332728   doi: 10.7527/S1000-6893.2025.32728

面向大规模在轨服务任务的快速重规划方法

徐杭1, 宋斌1,3, 刘鲁江2,3(), 李兴龙1, 袁秋帆1   

  1. 1.上海宇航系统工程研究所,上海 201109
    2.上海航天技术研究院,上海 201109
    3.宇航空间机构全国重点实验室,上海 201109
  • 收稿日期:2025-08-31 修回日期:2025-09-23 接受日期:2025-10-27 出版日期:2025-11-11 发布日期:2025-11-10
  • 通讯作者: 刘鲁江 E-mail:liulujiang24@163.com
  • 基金资助:
    中国航天科技集团有限公司青年拔尖人才支持工程资助项目;2023年东方英才计划青年项目(综合平台)

Fast replanning method for large-scale on-orbit servicing missions

Hang XU1, Bin SONG1,3, Lujiang LIU2,3(), Xinglong LI1, Qiufan YUAN1   

  1. 1.Shanghai Institute of Aerospace System Engineering,Shanghai 201109,China
    2.Shanghai Academy of Spaceflight Technology,Shanghai 201109,China
    3.National Key Laboratory of Aerospace Mechanism,Shanghai 201109,China
  • Received:2025-08-31 Revised:2025-09-23 Accepted:2025-10-27 Online:2025-11-11 Published:2025-11-10
  • Contact: Lujiang LIU E-mail:liulujiang24@163.com
  • Supported by:
    China Aerospace Science and Technology Corporation Youth Talent Support Program;the Youth Program of the Oriental Talents Plan in 2023 (Comprehensive Platform)

摘要:

针对大规模在轨服务任务规划需求,现有研究通常局限于静态场景,难以应对航天器故障、任务变更等突发情况导致的初始方案失效问题,工程应用需通过灵活的Lambert转移临机重规划。同时,该任务是包含轨迹规划、序列规划、任务分配3层嵌套优化的混合整数规划问题,现有策略迭代计算量大,尤其是针对需要数值迭代求解的Lambert转移,这严重制约了工程实际应用。针对这一难点,提出高效求解的算法框架:首先针对底层轨迹规划与序列规划问题,分别采用深度神经网络与注意力模型替代在线优化迭代过程,实现快速求解;其次,针对上层的任务分配问题,提出基于历史预案的集束预搜-邻域优化算法(BPNOA),通过目标分配状态标注、基于组合变轨策略的适配航天器粗筛、集束搜索与拍卖策略的预案调整、局部优化等机制,实现先快速寻优、后局部精调的高效求解过程。最后,以CelesTrak网站的真实轨道数据构建场景,将所提方法与遗传算法对比,小规模场景测试中本文方法收敛效率平均提高150多倍;大规模场景99次蒙特卡洛测试中,相同计算量下所提方法优化结果平均提高71%,多次求解标准差降低84.1%,收敛更优、更稳健。结果验证了新方法的有效性,该方法可显著提升任务重规划效率,满足未来应用的实时性需求。

关键词: 在轨服务, Lambert转移, 重规划, 在线优化, 集束搜索, 注意力模型

Abstract:

To meet the requirements of large-scale on-orbit service mission planning, existing research is usually confined to static scenarios, making it difficult to tackle the invalidation of initial plans caused by unexpected events such as spacecraft failures and task changes. This calls for on-the-fly replanning with flexible Lambert transfers. However, this mission is a mixed-integer programming problem involving three nested optimization layers: trajectory planning, sequence planning, and task assignment. Current strategies involve extensive iterative computations, especially for Lambert transfers that require numerical iterative solutions, which severely limits their practical engineering applications. To address this challenge, this paper proposes an efficient algorithmic framework. First, for the underlying trajectory planning and sequence planning problems, deep neural networks and attention models are adopted respectively to replace the online optimization iteration process, enabling rapid solutions. Second, for the upper-layer task assignment problem, a Beam Pre-search and Neighborhood Optimization Algorithm (BPNOA) based on historical pre-plans is proposed. By integrating mechanisms including target assignment state labeling, coarse screening of suitable spacecraft based on combined orbit transfer strategies, pre-plan adjustment using beam search and auction strategies, and local optimization, the proposed framework realizes an efficient solution process characterized by “fast initial optimization followed by fine-tuning”. Finally, using real orbital data from the CelesTrak website, comparative experiments with genetic algorithms demonstrate that in small-scale scenario tests, the proposed method exhibits an average convergence efficiency over 150 times higher; in 99 Monte Carlo tests of large-scale scenarios, with the same calculations, the average optimization result of the proposed method is improved by 71%, and the standard deviation of multiple solutions is reduced by 84.1%. The convergence is better and more robust. The results verify the effectiveness of the new method, which can significantly improve the efficiency of task replanning and meet the real-time requirements of future applications.

Key words: on-orbit service, Lambert transfer, re-planning, online optimization, beam search, attention model

中图分类号: