基于注意力Seq2Seq模型的终端区航空器航迹预测

为充分挖掘机场终端区航空器航迹时间依赖性,解决中长期、多步长航迹预测精度不稳定的问题,引入注意力机制(Attention Mechanism)和教师监督(Teacher Forcing)中的指数衰减(Exponential Decay)采样方法,提出了一种基于序列到序列框架的机场终端区航迹预测模型(Seq2Seq-Attention Mechanism-Exponential Decay, SAE)。序列到序列框架实现了多步长预测,注意力机制提高解码器预测精度,指数衰减采样方法加速了训练阶段模型收敛,在一定程度上提高了模型的泛化性。最后,为了验证提出方法的有效性,利用天津终端区28架次、90天ADS-B航迹数据构建原始数据集,以平均绝对误差(Mean Squared Error, MAE)和均方根误差(Root Mean Squared Error, RMSE)作为模型性能评价指标,进行了航迹预测实验,实验结果表明：高度、经度和纬度在序列到序列框架中的循环神经网络分别采用LSTM、GRU和LSTM可以获得最好预测性能;以四种预测长度1分钟、3分钟、5分钟和10分钟进行建模,与基线模型中预测性能最好的结果比较,所提出方法在验证集上的高度、经度和纬度指标表现最优,10分钟预测窗口下的平均绝对误差分别降低了66.30%, 54.62%和36.59%,均方根误差分别降低了65.45%, 38.16%和20.57%,同时,上述四种预测时长下所提出方法预测结果的均值和方差最小,表明随着预测时长的增加,模型预测结果的稳定性最好。此外,引入的注意力机制与指数衰减采样方法对有效捕捉航迹时间依赖性、提高模型泛化性均具有积极的贡献.

Aircraft Trajectory Prediction within Terminal Area Based on Seq2Seq-Attention Model

In order to fully explore the time dependence of aircraft trajectory within airport terminal areas and solve the problem of unstable medium and long term and multi-step trajectory prediction accuracy, an Attention Mechanism and Exponential Decay Sampling Method in Teacher Forcing are introduced, and a Seq2Seq-Attention Mechanism-Exponential Decay (SAE) model for airport terminal area trajectory prediction based on the Seq2Seq framework is proposed. The Seq2Seq framework achieves multi-step prediction, Attention Mechanism improves decoder prediction accuracy, Exponential Decay Sampling Method accelerates model convergence during the training phase, and to some extent improves model generalization. Finally, in order to verify the effectiveness of the proposed method, the original data set was constructed using the ADS-B trajectory data of 28 aircraft and 90 days in Tianjin Terminal Area, and the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) were used as model performance evaluation indicators to carry on aircraft trajectory prediction experiments. The experimental results show that the recurrent neural network of altitude, longitude and latitude within the Seq2Seq framework utilizing LSTM, GRU and LSTM can achieve the best predictive performance; Four prediction windows of 1 minute, 3 minutes, 5 minutes and 10 minutes were used for modeling. Compared with the results with the best prediction performance in the baseline model, the proposed method performed best in the altitude, longitude and latitude indicators on the verification set. MAE under the 10 minutes prediction window was reduced by 66.30%, 54.62% and 36.59%, and RMSE was reduced by 65.45%, 38.16% and 20.57%, respectively. At the same time, The mean and variance of the proposed method"s prediction results are the smallest under the four prediction durations mentioned above, indicating that as the prediction duration increases, the stability of the model"s prediction results is the best. In addition, the attention mechanism and exponential decay sampling method introduced have a positive contribution to effectively capturing trajectory time dependence and improving model generalization.