In order to further improve the accuracy of aircraft trajectory prediction,the encoder-decoder structure and multi-head self-attention mechanism are combined to construct a Transformer-based aircraft trajectory prediction model.This paper focuses on the data from the aircraft onboard quick access recorder(QAR)devices.The model training set and test set are divided in an 8:2 ratio.Trajectory features such as latitude,longitude,altitude,and speed are extracted to construct trajectory feature vectors,which are then input into the Transformer prediction model for training.The prediction results of the Transformer model,LSTM model,and BP model are evaluated and compared using MSE and the coefficient of determination(R2).The evaluation results demonstrate that the Transformer trajectory prediction model achieves higher prediction accuracy compared to the other two prediction models.
维普
万方数据