基于级联模糊系统的兵棋进攻关键点推理

Inference of Key Points of Attack in Wargame Based on Cascaded Fuzzy System

帮助兵棋AI学习兵棋推演中专家(人类指挥员)的知识和经验,有望提升其智能程度。在前期对兵棋专家知识进行分析归纳的过程中发现,专家知识中包含一类重要且使用频繁的隐性知识——作战任务规划关键点。以这些关键点为抓手,可以为兵棋AI的作战行动分配以及作战方案的制定增加可行性,进而提高其智能水平。以兵棋推演中的进攻作战任务为例,在对专家知识进行分析综合的基础上,利用级联模糊系统对模糊的态势信息进行推理,提取出进攻任务中的关键点。仿真结果表明采用的级联模糊推理系统可以较好地挑选出进攻任务中不同作战单元的关键点。

Helping wargame artificial intelligence(AI)to learn the knowledge and experience of the experts(human commanders)in wargaming is expected to improve the intelligence level of wargame AI.In the previous research of analyzing and summarizing the wargame expert knowledge,it is found that the expert knowledge contains a kind of important and frequently used tacit knowledge—key points of combat mission planning.Taking these key points as a starting point can increase the feasibility of operation assignment and operational plan formulation of wargame AI,thus improving its intelligence level.Taking the attack mission in wargaming for example,the cascade fuzzy system is used to infer the fuzzy battlefield situation information and extract the key points of the attack mission based on the analysis and synthesis of the expert knowledge.The simulation results show that the cascade fuzzy inference system can select the key points of different combat units in the attack mission.