面向随机并发任务的杀伤网建模与最优控制方法  

Modeling and Optimal Control of a Kill Web for Stochastic-oriented Concurrent Tasks

作　　者：吴克宇 冯旸赫 黄金才[1] 刘忠[1] WU Keyu;FENG Yanghe;HUANG Jincai;LIU Zhong(Key Laboratory of Big Data and Decision Making,National University of Defense Technology,Changsha Hunan 410073,China;School of Electronic Science,National University of Defense Technology,Changsha Hunan 410073,China)

机构地区：[1]国防科技大学大数据与决策重点实验室,湖南长沙410073 [2]国防科技大学电子科学学院,湖南长沙410073

出　　处：《指挥与控制学报》2023年第4期487-494,共8页Journal of Command and Control

基　　金：国家自然科学基金(62001495);中国博士后科学基金(48919)资助。

摘　　要：围绕随机并发任务场景,开展杀伤网建模和最优控制方法研究,提出了基于随机受控Petri网模型的杀伤网形式化建模方法,用以描述随机并发目标打击任务下的杀伤网的运行逻辑,以及在不同杀伤链选择动作下杀伤网状态的动态变化过程;针对最优杀伤链选择策略求解问题,提出了基于马尔科夫决策过程的最优策略构造范式和基于深度强化学习的最优策略求解方法.研究结果可为复杂对抗场景下的作战体系建模分析和优化控制的相关工作提供理论模型基础.Focusing on stochastic and concurrent task scenarios,the study on modeling and optimal control method of a kill web is carried out.Firstly,a formalized modeling of a kill web based on stochastic controlled Petri net is proposed,which depicts the operational logic of a kill web under stochastic and concurrent strike task and dynamic variation process of a kill web state under different kill chain selection actions.Further,for the solution problem of the optimal kill chain selection strategy,an optimal strategy solution method based on the optimal strategy construction paradigm of Markov decision-making process and deep reinforcement learning is proposed.The research result of this paper can provide theoretical model basis for the modeling and analysis of combat system under complex confrontation scenarios and the relative work of optimal control.

关 键 词：杀伤网 建模分析 优化控制 PETRI 网 强化学习 

分 类 号：E91[军事]