基于机器学习算法的海空目标航迹数据分析技术研究  

Study on Analysis of Sea Surface and Space Targets Trajectory Tracking Data Based on Machine Learning Algorithm

作　　者：李晓娜 杨焱煜 潘越[1] LI Xiao-na;YANG Yan-yu;PAN Yue(China Academy of Electronic Science,Beijing 100041,China)

机构地区：[1]中国电子科学研究院,北京100041

出　　处：《中国电子科学研究院学报》2020年第1期29-35,共7页Journal of China Academy of Electronics and Information Technology

摘　　要：文中以真实海面目标航迹数据和模拟的空中目标航迹数据为基础提出了一种基于机器学习算法的通用的海空目标航迹数据分析技术,为海空目标规律分析和航迹规划等提供理论支持。该技术通过对各类海空目标航迹数据的组成、航迹类别等特征进行分析,以整条航迹为研究对象,主要包含典型航迹生成方法、聚类航迹处理方法、航迹预测方法三个部分,通过DBSCAN算法对海空目标航迹进行聚类,并对聚类结果进行分段、压缩等操作以获取海空目标的拟合航迹,最后基于拟合航迹进行海空目标航迹预测。Based on machine learning algorithm,this article proposes an interchangeable sea surface and space targets track data analysis and mining technology based on real trajectory tracking data and simulate trajectory tracking data,thus providing theoretical support for law analysis of sea surface and space targets and trajectory planning. This technology analyzes the characteristics of various sea surface and space targets tracking data and categories,and includes three parts: typical tracking generation algorithm,cluster track processing method and tracking prediction method,and uses DBSCAN clustering algorithms to cluster the sea surface targets tracks. And then segments,compresses,fits the clustering results to obtain the classic trajectory of the sea surface and space targets,and finally to predict the trajectories of the sea surface and space targets.

关 键 词：海空目标 航迹 聚类 航迹拟合 

分 类 号：E11[军事—军事理论] TP18[自动化与计算机技术—控制理论与控制工程]