A Clustering Ensemble Method of Aircraft Trajectory Based on the Similarity Matrix

Performing clustering analysis on a large amount of historical trajectory data can obtain information such as frequent flight patterns aircraft and air traffic flow distribution, which provide reference for the revision standard procedures optimization division airspace sectors. At present, most uses single algorithm. When other processing remains unchanged, it is difficult to improve effect by using method. Therefore, this paper proposes ensemble method based similarity matrix. Firstly, stacked autoencoder used learn small number features that are sufficiently representative input subsequent Secondly, each basis cluster set, then consistent matrix obtained results cluster. On basis, deformation distance between trajectories, agglomerative hierarchical algorithm Taking Nanjing Lukou Airport terminal area an example, experimental show integrating multiple clusters eliminates inherent randomness algorithm, more robust.