基于多重聚类算法的分群化空战态势评估

Classifier Air Combat Situation Assessment Based on Multi-Reunion Class Algorithm

准确及时的作战态势分析是空战决策的首要前提,是飞行员空中飞行规划的主要依据,是夺取空战胜利的重要保证。针对空战过程数据信息量大,态势复杂多变的情况,依据军事实战需求,针对空中作战数据,提出了一种基于多重聚类算法进行态势分析的方法,解决了单机对单机作战中距离变化频繁、态势多变不易评估的难题。首先,采用K-means算法对获取的数据进行分群化处理,将整个空战过程按照敌我两机相对距离和状态分为不同群组;然后,根据每个群组数据特征采用不同的聚类算法进行态势分析,以使空战过程中的每个阶段都能达到最准确的态势分析结果;最后,利用飞行训练系统采集的真实数据进行仿真分析,结果表明,该方法具有良好的态势分析效果。

Accurate and timely combat situation analysis is the primary premise of air combat decisionmaking,the main basis of pilots air flight planning,and the important guarantee to win air combat.Aiming at the situation of large amount of data information and complex and changeable situation in air combat process,according to the requirements of military actual combat,a situation analysis method based on multi-reunion algorithm is proposed for air combat data,which solves the problem of frequent change of distance and difficult assessment of changeable situation in single-machine combat.First the K-means algorithm is adopted to obtain the data object,will the whole process of air combat according to the general state of two machine relative distance and divided into different groups,then according to the characters of each group of data using different clustering algorithm for situational analysis,in order to make each stage in the process of air can reach the most accurate analysis results,Finally,the real data collected by flight training system is used for simulation analysis,and the results show that this method has good situation analysis effect.