航空学报 > 2015, Vol. 36 Issue (11): 3652-3665   doi: 10.7527/S1000-6893.2015.0085

有人/无人机混合编队有限干预式协同决策

陈军, 张新伟, 徐嘉, 高晓光   

  1. 西北工业大学电子信息学院, 西安 710072
  • 收稿日期:2014-12-09 修回日期:2015-03-26 出版日期:2015-11-15 发布日期:2015-04-10
  • 通讯作者: 陈军,Tel.:029-88431260,E-mail:junchen@nwpu.edu.cn E-mail:junchen@nwpu.edu.cn
  • 作者简介:陈军,男,博士,副教授。主要研究方向:异构多飞行器编队的监督控制与协同决策。Tel:029-88431260,E-mail:junchen@nwpu.edu.cn;张新伟,男,硕士研究生。主要研究方向:模糊认知图理论及其在控制与决策中的应用。E-mail:npu_zhangxinwei@163.com
  • 基金资助:

    国家自然科学基金(61305133);航空科学基金(2013ZC53037);中央高校基本科研业务费专项资金(3102014JCY01005)

Human/unmanned-aerial-vehicle team collaborative decision-making with limited intervention

CHEN Jun, ZHANG Xinwei, XU Jia, GAO Xiaoguang   

  1. School of Electronics and Information, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2014-12-09 Revised:2015-03-26 Online:2015-11-15 Published:2015-04-10
  • Supported by:

    National Natural Science Foundation of China (61305133);Aeronautical Science Foundation of China (2013ZC53037);The Fundamental Research Funds for the Central Universities (3102014JCY01005)

摘要:

针对Leader-Follower异构型有人/无人机混合编队协同决策系统具有递阶分布式决策结构、决策信息分散以及通信约束等特点,在智能体模糊认知图(ABFCM)和动态模糊认知图(DFCM)的理论基础上,提出了一种有限干预式协同决策机制。通过构建层次化的Follower平台自主决策模型,实现了该平台与外部系统良好的交互能力,体现了自主决策的"动态性"。通过设计Leader平台的3种干预策略,满足了不同层次的决策需求,体现了干预过程的"有限性"。仿真结果表明:有限干预协同决策模型能够适应外部环境的动态变化,充分发挥Follower平台的自主决策能力;而不同层次的有限干预介入既减轻了Leader平台的控制负荷,又保证了决策的有效性和可行性,可为解决其他同类复杂系统的协同决策问题提供理论依据和方法参考。

关键词: 异构系统, 多智能体系统, 模糊推理, 协同, 决策, 模糊认知图, 有限干预

Abstract:

In view of the characteristics of Leader-Follower heterogeneous multi-platform system for the human/unmanned aerial vehicle team which are hierarchical and distributed structure, decision-making information decentralization and communication constraints, the collaborative decision-making mechanism with limited intervention based on the combination of agent-based fuzzy cognitive map (ABFCM) and dynamic fuzzy cognitive map (DFCM) is presented. In the proposed approach, hierarchical autonomous decision-making model for Follower platform is developed in order to make the Follower able to interact with external environment well and reflect the autonomous decision-making dynamics. Three intervention strategies under different conditions are designed for Leader platform to meet the requirements of different decision levels. Meanwhile these strategies reflect the limitation of intervention processes. Simulation results show that intervention-limited collaborative decision-making mechanism can adapt to dynamic changes in the external environment and make full use of Follower platform's autonomous decision-making ability. Hierarchical limited interventions can reduce the control workload of Leader and ensure the effectiveness and feasibility of decision-making as well. This work also provides a theoretical basis and methodology support to solve other similar collaborative decision-making problem in complex systems.

Key words: heterogeneous systems, multi-agent systems, fuzzy inference, collaboration, decision-making, fuzzy cognitive maps, limited intervention

中图分类号: