Trajectory clustering for arrival aircraft via new trajectory representation

Trajectory clustering can identify the flight patterns of air traffic, which in turn contributes to airspace planning, traffic flow management, and time estimation. This paper presents a semantic-based trajectory method for arrival aircraft via new proposed representation. The consists four significant steps: representing trajectories, grouping trajectories based on representation, measuring similarities between different through dynamic warping (DTW) each group, k-means density-based spatial applications with noise (DBSCAN). We take inbound toward Shanghai Pudong International Airport (ZSPD) carry out case studies. corresponding results indicate that could not only distinguish particular patterns, but also improve performance