面向多机协同空战的风格化分层策略设计方法  

A Stylized Hierarchical Strategy for Multi-aircraft Collaborative Air Combat

在线阅读下载全文

作  者:陈加乐 陈敏 冯锦元 蒲志强[2,3] 陈敏杰 CHEN Jiale;CHEN Min;FENG Jinyuan;PU Zhiqiang;CHEN Minjie(School of Computer Science,Nanjing University of Information Science&Technology,Nanjing 210044,China;Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China;School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100049,China;Huaru Research Institute,Beijing 100193,China)

机构地区:[1]南京信息工程大学计算机学院,南京210044 [2]中国科学院自动化研究所,北京100190 [3]中国科学院大学人工智能学院,北京100049 [4]华如研究院,北京100193

出  处:《指挥与控制学报》2025年第1期95-104,共10页Journal of Command and Control

摘  要:针对空战中局势动态变化、对手策略多样的难点,基于OODA层次化思想提出面向多机协同空战的风格化分层策略设计方法。orient阶段用参数表征风格,实现风格动态切换;decide阶段将协同决策问题建模为求解团队最大价值的优化问题,实现机群的协商与配合。实验表明,该方法对抗首届全国空中智能博弈大赛中6支高水平策略时平均胜率达到89.3%,且应对通信故障、武器失效等意外时呈现出极佳的鲁棒性,在对手策略变化的条件下展现出优异的泛化性。To address the challenges of dynamic changes in air combat situations and the diverse strategies of opponents,this study proposes a stylized hierarchical strategy design method for multi-aircraft cooperative air combat based on the OODA(observe,orient,decide,act)framework.In the orient phase,styles are characterized using parameters to enable dynamic style switching.In the decide phase,collaborative decision-making is modeled as an optimization task aimed at maximizing team value,facilitating fleet negotiation and cooperation.Experimental results demonstrate that the proposed method achieves an average winning rate of 89.3% against six high-level strategies in the first National Air Intelligence Gaming Competition.Additionally,the method exhibits excellent robustness in handling unexpected situations such as communication failures and weapon malfunctions,as well as strong generalization capabilities under varying opponent strategies.

关 键 词:多机协同空战 OODA 风格化 鲁棒性 泛化性 

分 类 号:TP18[自动化与计算机技术—控制理论与控制工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象