基于改进谱聚类的终端区航空器飞行轨迹分析

Aircraft Flight Trajectory Analysis in Terminal Area Based on Improved Spectral Clustering

为精确分析终端区空中交通流空间分布特征,有效评估空中交通管制服务水平,研究了基于自然邻的自适应谱聚类终端区飞行轨迹分布识别方法.在分析终端区飞行轨迹数据基础上,采用重采样方法保留飞行特征对航迹归一化处理,基于航向变化因子和速度变化因子建立相似度计算模型;通过自然邻算法获取航迹点间邻近信息,无需输入规模参数自适应实现高斯核函数降噪处理,采用改进谱聚类算法对终端区交通流聚类分析.以昆明长水机场进行实例验证.结果表明,方法能够有效识别终端区交通流分布.

In order to accurately analyze the spatial distribution characteristics of air traffic flow in terminal areas and effectively evaluate the service level of air traffic control,an adaptive spectral clustering method based on natural neighbors was studied to identify the distribution of flight trajectories in terminal areas.On the basis of analyzing the flight trajectory data in the terminal area,the resampling method was adopted to preserve the flight characteristics and normalize the flight trajectory,and a similarity calculation model was established based on the course change factor and the speed change factor.Neighbor information between track points was obtained through natural neighbor algorithm,and Gaussian kernel function noise reduction processing was realized adaptively without input of scale parameters.The improved spectral clustering algorithm was adopted for clustering analysis of traffic flow in terminal area.Taking Kunming Changshui Airport as an example,the results show that the method can effectively identify the traffic flow distribution in the terminal area.