基于多特征相似度的空中目标聚类研究  

Study on Air Targets Clustering Based on Multi-Features Similarity

作　　者：宁冰聪 范国梁[2] 罗恒钰[1] NING Bing-cong;FANG Guo-liang;LUO Heng-yu(China Academy of Electronics and Information Technology,Beijing 100041,China;Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China)

机构地区：[1]中国电子科学研究院,北京100041 [2]中国科学院自动化研究所,北京100190

出　　处：《中国电子科学研究院学报》2024年第5期480-486,共7页Journal of China Academy of Electronics and Information Technology

摘　　要：为准确识别空战场敌方目标的编队信息,本文提出一种针对空中目标的聚类模型,综合选取空中目标的位置、高度、航向、航速等多维度信息,计算目标之间的相似度,构建多特征相似度矩阵,将多特征相似度作为目标之间“距离”的度量方式,通过K-means算法聚类迭代,得到空中目标的编队信息。通过仿真实验,验证了本文所提方法在复杂空战场环境下,能准确识别敌目标的编队信息,相比传统聚类算法,识别准确率提高40%,有较好的工程应用价值。A model for clustering air targets is proposed to accurately identify the formation information of enemy targets in the air battlefield.This model comprehensively selects multi-feature such as the position、altitude、heading,and speed information of air targets to calculate the similarity between targets,and a multi-feature similarity matrix is constructed.Multi-feature similarity is used as a measure of“distance”between targets,and the formation information of air targets is obtained through clustering iteration using the K-means algorithm.Experimental results show that the identification accuracy is 40%higher than that based on traditional clustering algorithms,which proves the efficiency of the proposed method in this paper in solving identify formation information of enemy targets in complex air combat environments.

关 键 词：编队识别 聚类 特征相似度 K-MEANS 空中目标 

分 类 号：TN957.52[电子电信—信号与信息处理]