Aircraft Trajectory Prediction Based on Long and Short-Term Memory Structural Models

Collect ATC big data, according to the characteristics of ATC big data, perform data fusion on the data, process the trajectory data by using the improved clustering algorithm, and construct models for each type of obtained trajectory data separately to improve the prediction accuracy of the models. StackLSTM and convolutional LSTM-based aircraft trajectory prediction models are constructed respectively, and simulation experiments are carried out using real radar data as an example. The real radar data are used as an example for simulation experiments, and the simulation results are compared, and the results show that the aircraft trajectory prediction model based on convolutional LSTM can control the root-mean-square error of the prediction within the range of 0 to 1. The results show that the convolutional LSTM-based aircraft trajectory prediction model can control the root-mean-square error of the prediction within 400s, which verifies that the prediction model can realize the accurate prediction of aircraft trajectory.