基于模糊聚类的近距空战决策过程重构与评估

目前大量研究集中在空战的智能决策和解决近距空战评估结果"是什么"的问题上,却极少关注评估结果的"为什么"。根据空战训练中记录的客观数据的变化特征,提出基于模糊聚类的方法来计算决策序列,构建模糊粗糙决策系统,以实现对近距空战决策过程的重构;通过计算分析条件属性之间的相对重要度,对具有相似重要度的决策对象序列进行二次聚类划分,分析出关键决策点集合。通过实例研究,从空战能量和相对方位2个方面对近距空战决策过程进行了评估分析。结果表明,从空战决策的角度可认为关键决策点集合是产生评估结果的原因。

Reconstruction and evaluation of close air combat decision-making process based on fuzzy clustering

At present, a large number of researches focus on the area of intelligent decision-making and to solve the problem of what the evaluation results of air combat is, while the work of studying the reasons which lead to the evaluation results of air combat has received little attention so far. According to the changing characteristic of objective data recorded by air combat training system in air combat training, a fuzzy clustering method calculating the sequence of decision-making items is put forward, and then a fuzzy rough decision-making system is built to reconstruct the decision-making process of close air combat. By calculating and analyzing the relative importance degree between pairwise condition attributes, a two-phase clustering method is employed to deal with the decision-making items with similar importance. Then, the key decision-making sets are available. A case study is provided to analyze and evaluate the decision-making process of close air combat in terms of energy and relative position. Results show that the key decision-making sets result in the ultimate outcome of the air combat from the viewpoint of decision-making.