大语言模型驱动的多无人机协同任务规划(飞行器协同作战技术专栏)

  • 钟建鑫 ,
  • 孙景亮 ,
  • 罗野 ,
  • 李俊志 ,
  • 龙腾
展开
  • 1. 北京理工大学宇航学院
    2. 北京理工大学

收稿日期: 2026-03-24

  修回日期: 2026-10-21

  网络出版日期: 2026-10-23

基金资助

国家自然科学基金;北京市科技新星

LLM-Driven Collaborative Task Planning for Multi-UAV Systems

  • ZHONG Jian-Xin ,
  • SUN Jing-Liang ,
  • LUO Ye ,
  • LI Jun-Zhi ,
  • LONG Teng
Expand

Received date: 2026-03-24

  Revised date: 2026-10-21

  Online published: 2026-10-23

摘要

以大语言模型(Large Language Model, LLM)为代表的生成式人工智能具备任务语义理解与复杂推理等优势,在多无人机协同任务规划领域展现出巨大的发展潜力。然而,受限于上下文长度限制与规划知识图谱缺失,基于大语言模型的集群任务规划仍面临规划约束难以满足、任务执行意图易漂移以及动态环境适应性不足等问题。针对上述挑战,本文提出一种基于大语言模型驱动的多无人机协同规划方法,构建“顶层任务决策—态势感知分析—底层指令执行”多智能体协作框架。顶层结合领域知识库与任务规划工具,解析任务语义输入为图结构化原子任务序列,实现从抽象语义意图到可执行动作的高效映射;构建基于上下文管理的动态闭环执行机制,实时感知监控态势变化,采用增量调整策略应对突发任务,确保任务执行的长程逻辑一致性与动态适应性;设计“监督微调+强化学习”的大模型微调方案,提升模型在多无人机任务场景下的逻辑推理能力与决策一致性。仿真结果表明,相较于经典LLM自主决策框架ReAct,任务成功率提升了约30%,工具调用成功率达到95.2%,能够高效完成语义输入到多无人机自主协同任务执行的完整闭环。

本文引用格式

钟建鑫 , 孙景亮 , 罗野 , 李俊志 , 龙腾 . 大语言模型驱动的多无人机协同任务规划(飞行器协同作战技术专栏)[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2026.33604

Abstract

Large Language Models (LLMs) have demonstrated strong capabilities in semantic understanding and complex rea-soning, showing great potential for multi-UAV cooperative task planning. However, existing LLM-driven planning ap-proaches still face significant challenges, including constraint violations, semantic drift during long-horizon execution, and limited adaptability to dynamic environments due to insufficient structured planning mechanisms and constrained context capacity. To address these issues, this paper proposes an LLM-driven collaborative task planning framework for multi-UAV systems based on a multi-agent architecture. The framework integrates task decision, situational awareness analysis, and UAV-level execution into a unified cooperative process. High-level semantic instructions are transformed into graph-structured atomic task representations, enabling a structured mapping from abstract intentions to executable actions. A context-aware closed-loop mechanism is introduced to continuously monitor environmental changes and support incremental task adjustment under unexpected events, thereby maintaining logical consistency and improving dynamic responsiveness during long-horizon missions. Furthermore, a two-stage alignment strategy combining supervised fine-tuning and reinforcement learning is adopted to enhance the reasoning capability and deci-sion reliability of the LLM in multi-UAV task scenarios. Simulation results demonstrate that, compared with the classi-cal LLM-based planning framework ReAct, the proposed method improves task success rate by approximately 30% and achieves a 95.2% tool invocation success rate, effectively realizing a complete closed-loop process from semantic task input to autonomous multi-UAV cooperative execution.

参考文献

[1] 向锦武,董希旺,丁文锐,等. 复杂环境下无人集群系统自主协同关键技术[J]. 航空学报,2022, 43(10):333-365. [2] GARRETT C R, CHITNIS R, HOLLADAY R, et al. Integrated task and motion planning[J]. Annual review of control, robotics, and autonomous systems, 2021, 4(1):265-293. [3] GUO H, WU F, QIN Y, et al. Recent trends in task and motion planning for robotics:A survey[J]. ACM Computing Surveys, 2023, 55(13s):1-36. [4] 张宏宏, 李文华, 郑家毅, et al. 有人/无人机协同作战:概念, 技术与挑战[J]. 航空学报, 2023, 45(15):29653-029653. [5] AERONAUTIQUES C, HOWE A, KNOBLOCK C, et al. Pddl—the planning domain definition language[J]. Technical Report, Tech Rep, 1998. [6] FOX M, LONG D. PDDL2. 1:An extension to PDDL for expressing temporal planning domains[J]. Journal of artificial intelligence research, 2003, 20:61-124. [7] DE GIACOMO G, VARDI M Y. Linear temporal logic and linear dynamic logic on finite traces[C]//Proceedings of the 23rd International Joint Conference on Artificial Intelligence(IJCAI). AAAI Press, 2013:854-860. [8] JUVVI M S, KURNE T D, KOLATHAYA S, et al. Signal Temporal Logic Compliant Co-design of Planning and Control[C]//2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE Press, 2025:19204-19209. [9] BAIER C, KATOEN J-P. Principles of model checking [M]. MIT press, 2008. [10] TOUSSAINT M. Logic-Geometric Programming:An Optimization-Based Approach to Combined Task and Motion Planning[C]//Proceedings of the 24th International Joint Conference on Artificial Intelligence(IJCAI). AAAI Press, 2015:1930-1936. [11] TAKANO R, OYAMA H, YAMAKITA M. Continuous optimization-based task and motion planning with signal temporal logic specifications for sequential manipulation[C]//2021 IEEE international conference on robotics and automation (ICRA). IEEE Press, 2021:8409-8415. [12] XIAOLING S, DUAN L. Recent advances in integer programming[J]. Operations Research, 2014, 18(1). [13] WEI J, WANG X, SCHUURMANS D, et al. Chain-of-thought prompting elicits reasoning in large language models[J]. Advances in neural information processing systems, 2022, 35:24824-24837. [14] SINGH I, BLUKIS V, MOUSAVIAN A, et al. ProgPrompt:program generation for situated robot task planning using large language models[J]. Autonomous Robots, 2023, 47(8):999-1012. [15] HONG S, ZHUGE M, CHEN J, et al. MetaGPT:Meta programming for a multi-agent collaborative framework[C]//The 12th International Conference on Learning Representations. ICLR, 2023. [16] HU Z, LUCCHETTI F, SCHLESINGER C, et al. Deploying and evaluating llms to program service mobile robots[J]. IEEE Robotics and Automation Letters, 2024, 9(3):2853-2860. [17] ZITKOVICH B, YU T, XU S, et al. Rt-2:Vision-language-action models transfer web knowledge to robotic control[C]//Conference on Robot Learning (CoRL). PMLR, 2023:2165-2183. [18] WU Y, ZHU M, LI X, et al. VLA-AN:An Efficient and Onboard Vision-Language-Action Framework for Aerial Navigation in Complex Environments[J]. arXiv preprint arXiv:251215258, 2025. [19] GUO Z, XIA L, YU Y, et al. Lightrag:Simple and fast retrieval-augmented generation[J]. arXiv preprint arXiv:241005779, 2024. [20] 王祝,徐广通,龙腾. 基于定制内点法的多无人机协同轨迹规划[J]. 自动化学报,2023,49(11):2374-2385. [21] LUO H, LIU Y, ZHANG R, et al. Toward edge general intelligence with multiple-large language model (Multi-LLM):architecture, trust, and orchestration[J]. IEEE Transactions on Cognitive Communications and Networking, 2025. [22] HU E J, SHEN Y, WALLIS P, et al. Lora:Low-rank adaptation of large language models[C]//The 11th international conference on learning representations. ICLR, 2022:3. [23] GUO D, YANG D, ZHANG H, et al. Deepseek-r1:Incentivizing reasoning capability in llms via reinforcement learning[J]. arXiv preprint arXiv:250112948, 2025. [24] SCHULMAN J, WOLSKI F, DHARIWAL P, et al. Proximal policy optimization algorithms[J]. arXiv preprint arXiv:170706347, 2017. [25] YANG A, LI A, YANG B, et al. Qwen3 technical report[J]. arXiv preprint arXiv:250509388, 2025. [26] GLM T, ZENG A, XU B, et al. Chatglm:A family of large language models from glm-130b to glm-4 all tools[J]. arXiv preprint arXiv:240612793, 2024. [27] AGARWAL S, AHMAD L, AI J, et al. gpt-oss-120b & gpt-oss-20b model card[J]. arXiv preprint arXiv:250810925, 2025. [28] YAO S, ZHAO J, YU D, et al. React:Synergizing reasoning and acting in language models[C]//The 11th international conference on learning representations(ICLR). ICLR, 2022.
Options
文章导航

/

〈 〉