基于ADS-B的4D航迹预测及应用

硕士学位论文基于ADS-B的4D航迹预测及应用研究生姓名杨东玲导师姓名焦卫东副教授2017年5月22日分类号 TP391.4 密级公开 UDC 621.39 学号 1404054 中国民航大学硕士学位论文基于ADS-B的4D航迹预测及应用研究生姓名杨东玲导师姓名焦卫东副教授申请学位类别工程硕士专业学位学科专业名称电子与通信工程所在院系电子信息与自动化学院论文答辩日期2017年5月14日2017年5月22日4D Track Prediction Based on ADS-B and ApplicationA Dissertation ted to Civil Aviation University of ChinaFor the Professional Degree of Master of EngineeringBYYang Dong-lingSupervised byProf. JIAO Wei-dongCollege of Electronic Ination and AutomationCivil Aviation University of China May 2017摘要作为新一代航空自动化系统的核心技术之一，航迹预测可用于冲突检测与避撞、飞行流量预测、进离场排序等方面，是保证航空运输高效性和安全性的一项关键技术。我国在航迹预测方面的研究相对欧美国家还处于追赶阶段，亟需准确有效的航迹预测方法。因为ADS-B具备数据信息丰富、更新速度快等优点，本文对基于ADS-B的4D航迹预测进行了系统深入的研究。首先，介绍了各种常用的航迹预测算法，并用ADS-B数据对各种预测算法进行仿真、对比，分析各种预测算法的优缺点及应用范围。然后，根据对比结果选出较优的算法进行整合改进，提出了一种基于自适应时空航迹聚类的航迹预测方法。这种方法的主要思想是将飞机运动分为直线运动阶段和转弯阶段，在直线运动阶段选用耗时较短的低分辨率聚类，在转弯阶段采用精度较高的高分辨率聚类。对飞机不同的运动阶段自适应地选用不同聚类方法，可以在保证一定精度的前提下有效减小运行时间。对通过自适应的时空航迹聚类算法获得的典型航迹进行当前位置修正，即可得到预测航迹。最后，通过探究空中交通流量的预测方法，说明了航迹预测的应用方向。并将基于自适应时空航迹聚类的航迹预测成果在U-GIS三维仿真平台上直观的显示。关键词 ADS-B；航迹预测；自适应时空航迹聚类；航迹预测应用AbstractWith the rapid development of aviation transportation industry, the contradiction between the sharp increase of traffic flow and limited airspace resources becomes more and more remarkable. As one of the core technologies of the new generation aviation automation system, trajectory prediction is a key technology that can ensure efficiency and safety of air transport, which can be used for collision detection and collision avoidance, flight flow detection, etc. Compared with the Europe and America, our country is still in the after status when it comes to track prediction research. So effective track prediction needs to be come up.Based upon the conditions described above, 4D track prediction is in-depth researched based on ADS-B data, under the premise that ADS-B data contains many advantages such as rich ination, faster update rate, easier to be get and so on. First of all, the various algorithms for track forecast are introduced, simulated, analysed, and compared using ADS-B dates in order to learn their Advantages and disadvantages and application scopes. Then, Edge algorithm is elected and improved so that an adaptive space-time track clustering algorithm is proposed used for track forecast. Flight paths are be divided into linear motion and turning stages.Different clustering algorithms in resolution are choosed adaptively depending on different stages of movement that can guarantee accuracy and time efficiency to a certain extent. Finally, the forecast of air traffic flow is explored, which is one of the applications of trajectory prediction based on ADS-B. And the research results are displayed directly on the U-GIS plat.Keywords ADS-B；Trajectorty Prediction；Adaptive Spatio-temporal Trajectory Cluster; Application of Trajectorty Prediction目录摘要IAbstractII第一章绪论11.1 课题研究背景11.2 国内外研究现状21.3 主要工作及论文结构3第二章 ADS-B数据选取与预处理52.1 ADS-B数据准备52.2 ADS-B数据预处理62.2.1 排除干扰航迹62.2.2 数据缺失航迹62.2.3重复数据点剔除72.3 坐标转换72.4 本章小结8第三章 4D航迹预测算法93.1 主流4D航迹预测算法93.1.1 基于灰色理论的航迹预测93.1.2 基于大圆/等角航线的航迹预测103.1.3 基于飞行意图的航迹预测113.1.4 基于数据挖掘的航迹预测133.3 仿真实验153.3 仿真结果对比223.4 本章小结23第四章基于自适应的时空航迹聚类及预测244.1 时空航迹数据244.2 时空航迹聚类及预测254.2.1 时间K-mea