LSTM-Cubic A*-based auxiliary decision support system in air traffic management

Abstract

Airports, pilots and geographical environment contribute a lot to stable and safe air traffic, especially during the climbing and descending phases. Accidents occur frequently in these two phases, even small errors can lead to serious consequences. Modern equipment has made great progress in monitoring the airspace. However, when encountering strong external interference or blind zones, potential risks will be raised. With the assistance of newly developed technologies and equipment, flight trajectories can be recorded at high frequency, which enhances capabilities of Air Traffic Management systems. A four-dimensional flight trajectory prediction model is put forward in this paper. Combined with sliding windows, Long Short-Term Memory network maintains the long-term features and manages to predict accurate trajectories. Massive data contributes a lot for forecasting, but cannot guarantee better decisions. Decision-making is closely related to threat assessment. Taking the geographical environment into consideration, the static threat is involved in the current work. The experimental scenario is set to be an aircraft takes off from Hong Kong International Airport, which is located close to mountains. Characteristics of terrains are introduced to our system by digital elevation model. The proposed cubic A* search algorithm provides reasonable path by considering the ultimate dynamic performance of an aircraft. Threat is assessed during the path planning process. Finally, the auxiliary decision support system is developed based on ArcGIS 10.0, to graphically provide the intuitive and quick assistance. Multiple sets of experimental results indicate that this system is able to provide timely decision support.