Trajectory Clustering-Based Controller Preference Identification Under Convective Weather

Abstract:
In order to accurately analyze controller preference under sequential convective weather, the paper proposes a trajectory clustering-based controller intention identification method for convective weather scenes. Firstly, the approach flight trajectories are resampled by using the uniform parameterization method; then the approach trajectories are clustered by using DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, and the central trajectories of traffic flows are identified by using K-means algorithm; then the trajectory clustering results are combined with the convective weather images to extract controller intention. Finally, the paper carries out case studies on typical scenes in the terminal area of Guangzhou. The results show that the proposed method can accurately identify controller intention under sequential convective weather. The results verify the effectiveness of the method.