摘要
随着航天技术向智能化、集群化发展,空间无人系统在深空探测、对地观测等战略领域展现出巨大潜力,但传统集中式控制范式在应对高动态环境、分布式任务和严格资源约束时面临严峻挑战.多智能体强化学习以其分布式决策架构和协同演化机制,为构建自主、弹性的空间智能系统提供了突破性解决方案.本文系统探讨了多智能体强化学习在空间无人系统中的技术赋能路径、方法体系、工程挑战与发展机遇;剖析了卫星集群协同通信和多航天器控制等核心场景的技术瓶颈;总结了空间无人系统在上述核心场景中的研究与应用现状;展望了多智能体强化学习作为新兴智能技术,在动态频谱分配、星载边缘计算和抗扰协同控制等关键方向的应用前景,推动空间系统向“自主决策-弹性抗扰-高效协同”的新范式演进.本文旨在为构建新一代空间智能无人集群提供现有技术梳理与前景展望.
With the advancement of space technology towards intelligence and clusterisation,unmanned space systems demonstrate immense potential in strategic areas such as deep space exploration and Earth observation.However,traditional centralized control paradigms face significant challenges in adressing highly dynamic environments,distributed tasks,and strict resource constraints.Leveraging its distributed decision-making architecture and co-evolutionary mechanisms,multi-agent reinforcement learning(MARL)offers a breakthrough solution for building autonomous and resilient intelligent space systems.This paper systematically explores MARL’s technological empowerment pathways,methodologies,engineering challenges,and opportunities in unmanned space systems.It analyzes the technical bottlenecks in core scenarios(e.g.,collaborative communication for satellite clusters,multi-spacecraft control).Moreover,It reveals the application mechanisms of MARL in critical domains,including dynamic spectrum allocation,on-board edge computing,and robust collaborative control.Finally,the paper proposes an integrated collaborative intelligence architecture that incorporates space-dynamics constraints with innovative MARL algorithms.This framework aims to drive the evolution of space systems toward a new paradigm of autonomous decision-making,resilient anti-jamming capabilities,and efficient collaboration.This research seeks to provide theoretical support and a technological roadmap for the next-generation space-based intelligent networks.
作者
李勐
冯肇晗
梅云鹏
曹宏杰
张博
王钢
LI Meng;FENG Zhaohan;MEI Yunpeng;CAO Hongjie;ZHANG Bo;WANG Gang(Beijing Institute of Technology,Beijing 100081,China;China Academy of Electronics and Information Technology,China Electronics Technology Group Corporation,Beijing 100041,China)
出处
《空间控制技术与应用》
北大核心
2025年第4期17-28,共12页
Aerospace Control and Application
基金
国家自然科学基金资助项目(U23B2059)。
关键词
多智能体强化学习
空间无人系统
协同控制
边缘计算
自主决策
multi-agent reinforcement learning(MARL)
unmanned space systems
collaborative control
edge computing
autonomous decision-making
