Aircraft Surface Trajectory Prediction Method Based on LSTM with Attenuated Memory Window

Abstract

With the rapid development of the global air transport industry, airport surface traffic is increasingly busy. The safety hazards of taxiing conflict still exist during the airport operation for the initially planned taxiing path about aircraft, which directly affects the operational safety and efficiency of the airport surface. Using the correlation and dependence between the position sequences of aircraft glide motion, a method based on Long Short Term Memory networks (LSTM) is used. Combining the change of the motion state of the surface aircraft, the attenuation memory window is introduced to improve the hidden layer structure to further enhance the prediction accuracy in the LSTM model, and it is compared and verified under different parameters. The method realizes the target of surface position prediction for the aircraft in future period, lays the foundation for aircraft surface taxiing conflict detecting, and avoids taxiing conflict.