Aircraft approach risk identification based on cluster analysis of flight surveillance data in terminal area

In order to ensure the safety of aircraft flights in the terminal area and to strengthen air traffic management, a research method based on trajectory clustering is used to make risk identification of abnormal trajectories in the terminal area. Firstly, the distance function of clustering is selected, and the dynamic time warping method is used to cluster and make trajectory anomaly detection for the approach trajectories in the terminal area and to depict out the trajectory similarity. Then the risk assessment is achieved by calculating the length of the distance from the abnormal trajectories to the center of the clusters. Finally, experimental results based on aircraft approach surveillance data in actual terminal areas show that dynamic time warping achieves effective results and successfully identifies various types of risks.