基于改进K-means算法的陆战场机动目标分群方法  

A clustering method of land battlefield maneuvering targets based on improved K-means algorithm

作　　者：张绪亮[1] 张宏军 綦秀利[1] 王文博[1] 王泽[1] 尹成祥[1] 

机构地区：[1]解放军理工大学,南京210007

出　　处：《信息技术》2016年第3期128-131,135,共5页Information Technology

摘　　要：陆战场上的多个机动兵力集群在某些情况下可能出现队形相互交错的情况,运用传统的基于实体空间信息的聚类算法(如K-means算法)进行实体目标分群时效果不理想。通过对实体空间信息进行分析与处理,提出了一种基于改进的K-means算法的陆战场机动目标分群方法,重新设计了用于K-means算法的距离度量函数,引入模块度的概念,实现了k值的自适应选取,通过实验验证了该方法对呈较复杂分布的陆战场机动目标进行分群的合理性。The formation of a plurality of motor strength cluster of land battlefield may stagger in some cases. The effect is not ideal that using traditional clustering algorithms based on spatial information（ such as K-means algorithm） to target entity clustering. Through the analysis and processing of spatial information,this paper proposes a clustering method of land battlefield maneuvering targets based on improved K-means algorithm,by redesigning the distance measure function and introducing the concept of modularity to achieve the adaptive selection of the value of k. The rationality of using this method in classifying the complexly distributed maneuvering targets on land battlefield is proved through the experiment.

关 键 词：态势估计 目标分群 K-MEANS算法 模块度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]