多头图注意力驱动特征聚合的巨型遥感星座任务规划算法
Multi-head graph-attention driven feature aggregation for large-scale satellite mission planning
- 电信科学 2026年42卷 页码：1-14
- 作者机构：
1.华中科技大学网络空间安全学院第六代移动通信研究中心，湖北 武汉 430074
2.西安电子科技大学电子工程学院，陕西 西安 710071
- 作者简介：
[ "佘雨璇（2002- ），男，华中科技大学网络空间安全学院第六代移动通信研究中心硕士生，主要研究方向为卫星互联网。" ]
[ "周康燕（1981- ），男，西安电子科技大学电子工程学院在读博士，主要研究方向为卫星互联网。" ]
[ "代奥（2001- ），男，华中科技大学网络空间安全学院第六代移动通信研究中心硕士生，主要研究方向为卫星互联网。" ]
[ "魏子翔（2002- ），男，华中科技大学网络空间安全学院第六代移动通信研究中心硕士，主要研究方向为卫星互联网。" ]
[ "周家喜（1980- ），男，华中科技大学网络空间安全学院第六代移动通信研究中心教授，主要研究方向为卫星网络智能任务规划、智能资源调度与星载智能体安全等。" ]
[ "肖丽霞（1987- ），女，华中科技大学网络空间安全学院第六代移动通信研究中心研究员，主要研究方向为卫星网络波形设计与智能资源规划。" ]
- 基金信息：国家自然科学基金资助项目(62571202)
DOI：10.11959/j.issn.1000-0801.2026119
中图分类号： TN927.2;TP393收稿：2025-12-02，
修回：2026-01-27，
录用：2026-01-28，
网络首发：2026-03-30，
纸质出版：2026-04-20
- 稿件说明：
移动端阅览
佘雨璇,周康燕,代奥等.多头图注意力驱动特征聚合的巨型遥感星座任务规划算法[J].电信科学,
She Yuxuan,Zhou Kangyan,Dai Ao,et al.Multi-head graph-attention driven feature aggregation for large-scale satellite mission planning[J].Telecommunications Science,
佘雨璇,周康燕,代奥等.多头图注意力驱动特征聚合的巨型遥感星座任务规划算法[J].电信科学, DOI：10.11959/j.issn.1000−0801.2026119.
She Yuxuan,Zhou Kangyan,Dai Ao,et al.Multi-head graph-attention driven feature aggregation for large-scale satellite mission planning[J].Telecommunications Science, DOI：10.11959/j.issn.1000−0801.2026119.
摘要
遥感卫星星座规模的持续扩大，极大地增强了其在资源普查、环境监测和灾害应急等领域的综合应用能力。然而，星座规模增长也显著增加了任务规划的复杂性。多智能体强化学习是解决大规模星座任务规划的有效途径，但卫星数量激增也带来智能体状态空间维度爆炸难题。为此，提出一种多头图注意力驱动特征聚合（multi-head graph attention driven feature aggregation，MGADFA）的巨型遥感星座任务规划算法。该方法利用多头图注意力捕捉卫星间的动态交互权重，通过特征聚合机制将复杂的多体交互转化为个体与群体的关联，在保留协作特征的同时，将网络参数规模从指数级降为线性级。仿真结果表明，所提算法在任务接受量和负载均衡性等方面均表现更优。
Abstract
The continuous expansion of remote-sensing satellite constellations has significantly enhanced their comprehensive application capabilities in resource surveys
environmental monitoring
and emergency disaster response. However
the burgeoning scale of these constellations markedly increases the complexity of mission planning. While multi-agent reinforcement learning is an effective approach for large-scale mission planning
the surge in satellite numbers leads to the challenge of dimensionality explosion in the state space. To address this bottleneck
a multi-head graph attention driven feature aggregation (MGADFA) algorithm for mission planning in giant remote-sensing constellations was proposed. This method utilized multi-head graph-attention to capture dynamic interaction weights between satellites and employed a feature aggregation mechanism to transform complex multi-body interactions into individual-population associations. While preserving key cooperative features
this approach reduced the joint decision space from an exponential scale to a linear dimension. Simulation results demonstrated that the proposed algorithm exhibited superior performance in terms of task throughput and load balancing.
关键词
Keywords
references
胡笑旋 , 王执龙 , 夏维 , 等 . 遥感卫星任务规划技术: 现状与展望 [J ] . 指挥与控制学报 , 2023 , 9 ( 5 ): 495 - 507 .
Hu X X , Wang Z L , Xia W , et al . Mission planning technology for remote sensing satellite: status and prospect [J ] . Journal of Command and Control , 2023 , 9 ( 5 ): 495 - 507 .
Peng G S , Song G P , Xing L N , et al . An exact algorithm for agile earth observation satellite scheduling with time-dependent profits [J ] . Computers & Operations Research , 2020 , 120 : 104946 .
Zeng G M , Zhan Y F , Xie H R , et al . Resource allocation for networked telemetry system of mega LEO satellite constellations [J ] . IEEE Transactions on Communications , 2022 , 70 ( 12 ): 8215 - 8228 .
Wei L N , Chen M , Xing L N , et al . Knowledge-transfer based genetic programming algorithm for multi-objective dynamic agile earth observation satellite scheduling problem [J ] . Swarm and Evolutionary Computation , 2024 , 85 : 101460 .
张泽华 , 张加友 , 张嘉凯 , 等 . 基于遗传禁忌算法的多星协同任务规划方法 [J ] . 无线电工程 , 2022 , 52 ( 7 ): 1127 - 1135 .
Zhang Z H , Zhang J Y , Zhang J K , et al . Multi-satellite cooperative mission planning method based on genetic tabu algorithm [J ] . Radio Engineering , 2022 , 52 ( 7 ): 1127 - 1135 .
Chen Y , Zhang D Y , Zhou M Q , et al . Multi-satellite observation scheduling algorithm based on hybrid genetic particle swarm optimization [M ] // Advances in Information Technology and Industry Applications . Berlin, HeidelbergSpringer 2012 : 441 - 448 .
Gu Y , Han C , Chen Y H , et al . Large region targets observation scheduling by multiple satellites using resampling particle swarm optimization [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2023 , 59 ( 2 ): 1800 - 1815 .
Han C , Gu Y , Wu G H , et al . Simulated annealing-based heuristic for multiple agile satellites scheduling under cloud coverage uncertainty [J ] . IEEE Transactions on Systems, Man, and Cybernetics: Systems , 2023 , 53 ( 5 ): 2863 - 2874 .
Khojah H A , Mosa M A . Multi-objective optimization for multi-satellite scheduling task: Multi-objective optimization for multi-satellite scheduling task [J ] . Journal of Soft Computing Exploration , 2022 , 3 ( 1 ): 19 - 30 .
Liu Y C , Chen Q F , Li C Y , et al . Mission planning for Earth observation satellite with competitive learning strategy [J ] . Aerospace Science and Technology , 2021 , 118 : 107047 .
Chen X Y , Tian T , Dai G M , et al . Deep reinforcement learning-based resource allocation method for multi-satellite scheduling [J ] . Computers & Operations Research , 2025 , 181 : 107088 .
Ou J W , Xing L N , Yao F , et al . Deep reinforcement learning method for satellite range scheduling problem [J ] . Swarm and Evolutionary Computation , 2023 , 77 : 101233 .
李英玉 , 史好迎 , 赵通 . 面向即时响应的卫星在轨分布式协商智能任务规划 [J ] . 空间科学学报 , 2024 , 44 ( 1 ): 159 - 168 .
Li Y Y , Shi H Y , Zhao T . On-orbit distributed negotiation intelligent mission planning for instant response [J ] . Chinese Journal of Space Science , 2024 , 44 ( 1 ): 159 - 168 .
Herrmann A , Stephenson M A , Schaub H . Single-agent reinforcement learning for scalable earth-observing satellite constellation operations [J ] . Journal of Spacecraft and Rockets , 2024 , 61 ( 1 ): 114 - 132 .
Zhang G H , Li X H , Hu G X , et al . MARL-based multi-satellite intelligent task planning method [J ] . IEEE Access , 2023 , 11 : 135517 - 135528 .
Lowe R , Wu Y , Tamar A , et al . Multi-agent actor-critic for mixed cooperative-competitive environments [C ] // Proceedings of the 31st International Conference on Neural Information Processing Systems . New York : ACM , 2017 : 6382 - 6393 .
Liu S , Chen Y W , Xing L N , et al . Time-dependent autonomous task planning of agile imaging satellites [J ] . Journal of Intelligent & Fuzzy Systems , 2016 , 31 ( 3 ): 1365 - 1375 .
Liu Z , Gao L , Chai Y . Research on task scheduling mechanism of distributed satellite system [J ] . Journal of Computational Information Systems , 2014 , 10 ( 24 ): 10703 - 10713 .
Nowé A , Vrancx P , De Hauwere Y M . Game theory and multi-agent reinforcement learning [M ] //Wiering M, van Otterlo M. Reinforcement Learning: State-of-the-Art . Berlin, Heidelberg : Springer , 2012 : : 441 - 470 .
Veličković P , Cucurull G , Casanova A , et al . Graph attention networks [C ] // Proceedings of the International Conference on Learning Representations (ICLR) . 2018 .
Zhang S , Tong H H , Xu J J , et al . Graph convolutional networks: a comprehensive review [J ] . Computational Social Networks , 2019 , 6 : 11 .
Mnih V , Kavukcuoglu K , Silver D , et al . Human-level control through deep reinforcement learning [J ] . Nature , 2015 , 518 ( 7540 ): 529 - 533 .
Yu C , Velu A , Vinitsky E , et al . The surprising effectiveness of PPO in cooperative multi-agent games [C ] // Proceedings of the 36th International Conference on Neural Information Processing Systems . New York : ACM , 2022 : 24611 - 24624 .
Abdi H . Coefficient of variation [J ] . Encyclopedia of Research Design , 2010 , 1 ( 5 ): 169 - 171 .
0
浏览量
0
下载量
0
CSCD
- 网络PDF下载
- WORD下载
- XML下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies
- 总访问量：932431 今日访问量：799