论文部分内容阅读
战场范围扩大、战斗实体位置疏散使得按照位置进行聚类的方法可信性降低。担负同一作战任务的若干战斗实体往往因保持队形的需要而具有相似的运动规律,因此,使用实体位置、速度、加速度等运动参量进行模糊聚类能够克服位置疏散带来的聚类困难。对实体的运动参量进行归一化,建立模糊等价关系矩阵,根据需要选择适当的截集水平进行聚类,得到不同级别建制的聚类结果。经实验数据检验,该方法能够取得满意的聚类效果,辅助各级指挥员对战场态势进行分析和评估。
The expansion of the battlefield and the evacuation of combat entities have reduced the credibility of the method of clustering by location. Some combat entities that are responsible for the same combat mission often have similar motion rules because of the need of team formation. Therefore, fuzzy clustering using physical parameters such as position, velocity and acceleration can overcome the difficulty of cluster evacuation. The physical parameters of the entity are normalized, and the fuzzy equivalent relation matrix is established. According to the requirement, an appropriate cut-off level is selected for clustering, and the clustering results of different levels are obtained. The experimental data show that this method can achieve satisfactory clustering effect and assist commanders at all levels to analyze and assess the battlefield situation.