This paper proposes an air traffic clustering approach leveraging Procrustes transformation for feature extraction, such as Procrustes distance, scale, rotation and translation. Features are weighted to balance their units and importance within a two-layer clustering approach. We compare the performance of our method to a well-accepted air traffic clustering technique including a more comprehensive evaluation of performance trade-offs, considering outliers, number and quality of the clusters. Evaluations encompass multiple airport regions, changes in database size and data resampling. The results demonstrate the robustness and effectiveness of our approach in addressing the challenges associated with increasing air traffic, heterogeneous air traffic, and uncrewed flights accessing the airspace.
Air Traffic Trajectory Clustering Using Procrustes Analysis
2023-10-01
2211929 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Stratified Generalized Procrustes Analysis
British Library Online Contents | 2013
|British Library Online Contents | 2017
|L 1 Generalized Procrustes 2D Shape Alignment
British Library Online Contents | 2008
|Passive Three Dimensional Face Recognition Using Iso-Geodesic Contours and Procrustes Analysis
British Library Online Contents | 2013
|Feature Matching with Procrustes Alignment and Graph Editing
British Library Conference Proceedings | 1999
|