基于聚类思想的战场群目标识别方法  

Group of Targets Recognition Method in Battlefield Based on Cluster Thought

作　　者：李全龙[1] 刘洪娟[1] 余硌[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《计算机工程》2007年第15期193-195,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60475012)

摘　　要：面向现代化战争的快速决策的需求,提出了一个基于聚类思想的战场群目标识别算法。该算法依据目标的位置、速度、行进方向等特征量,通过相异度分析的聚类方法对战场目标进行目标群识别。该算法时间复杂度低、识别准确率较高、对奇异点的处理更加健壮,能够满足战场实时性需求。仿真实验证明,该算法能以较大的概率准确地识别出战场目标群,为快速作战决策和战术规划提供支持。To meet the fast decision-making of modem warfare, an algorithm for group of targets recognition in battlefield is proposed, which is based on cluster thought. By cluster method of dissimilarity analysis, this algorithm makes use of the characteristic measure of location, speed and direction of targets to recognize groups of targets in battlefield. The advantage of the algorithm is its low complexity, much higher recognition accuracy rate and much more robust handling for outlier. The algorithm can satisfy the real time requirement in battlefield. Simulation experiments demonstrate that the algorithm can accurately recognize the group of targets in battlefield with higher probability and provide direct support for fast decision-making of battle and programming of tactics.

关 键 词：群目标 识别 聚类分析 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]