Aircraft Trajectory Clustering in Terminal Area Based on HDBSCAN Algorithm
Aiming at the problem of complex operating environment and uneven distribution of air traffic flow den-sity in the airport terminal area,a Fast Dynamic Time Warping(FastDTW)combined with Hierarchical Density-Based Spatial Clustering of Applications with Noise(HDBSCAN)algorithm is proposed to identify traffic flows with density differences in order to explore the traffic flow patterns in the terminal airspace.Combining the multidimension-al characteristics of the trajectory,the original trajectory data are resampled;the trajectory similarity matrix is con-structed using the FastDTW method,and the HDBSCAN method is applied to cluster the input matrix.The simulation results show that the above method can achieve fine clustering of trajectories and effectively identify air traffic flows of different densities by using real trajectory data in the terminal area for simulation verification.
Air transportationAircraft trajectory clusteringDynamic time warpingHDBSCAN
NETL
NSTL
万方数据
维普
