分布式智能体无人机集群的上下文与通信管理
原标题：Managing Context and Communication in Distributed Agentic UAV Swarms
AI 摘要
该论文提出一种全分布式无人机集群架构，每架无人机搭载独立的小语言模型（SLM）智能体、上下文管理器和兴趣感知 Gossip 引擎，通过事件驱动的 reason-act-observe 生命周期实现本地持续控制。运行时知识被表示为结构化原子笔记，并组织为核心、本地和同伴专属记忆，由确定性兴趣感知 Gossip 引擎根据接收方语义新颖性和时效性选择性传播。在十架无人机的模拟搜救任务中，该方法完成全部实验，而无限制洪泛仅完成 70–85%，将转发决策交给 SLM 则每次均无法完成任务；相比洪泛，该方法推理 token 消耗约减半、传输数据减少且幸存者计数误差更低。
正文节选
Managing Context and Communication in Distributed Agentic UAV SwarmsThanks: This paper has been accepted for presentation at the 24th IEEE Consumer Communications & Networking Conference 2027 (CCNC 2027) Abstract Unmanned aerial vehicle (UAV) swarms increasingly rely on language-model agents to provide adaptive mission-level reasoning in uncertain environments. Fully distributed control, in which each UAV hosts an independent Small Language Model (SLM), removes reliance on a centralized coordina