摘要
基于数学模型或统计模型的传统航迹预测方法存在一定的局限性,无法满足现代航空领域对于高效、准确、实时的航迹预测需求。针对此问题,提出基于注意力机制的CNN-LSTM模型的实时航迹预测方法。该模型首先使用一维卷积对航迹数据的多维度特征进行提取,从而减少输入特征的数量。其次利用获取的多维度时序数据作为LSTM的输入,通过LSTM提取上下文的信息。最后使用注意力机制为LSTM中不同时序节点的输出赋予权重,达到聚焦关键航迹信息的作用。经过实验验证:本文的模型与LSTM模型和CNN-LSTM模型相比,预测出的路径更接近真实航迹;文中的模型比LSTM模型的平均预测误差降低了29.7%,比CNN-LSTM模型降低了25.4%。综上所述,文中方法可以显著提高航迹预测的精度。
Aimed at the problems that traditional trajectory prediction methods based on mathematical or statistical models have a certain of inherent limitations and are difficult to meet increasingly the demands of efficiency,accuracy,and real-time trajectory prediction in the modern aviation field,a novel real-time traj-ectory prediction method is proposed based on a CNN-LSTM model with an attention mechanism.The proposed model is that multidimensional features are extracted from trajectory data by one-dimensional convolution,reducing the number of input features.Taking the resulting multidimensional time-series da-ta as an input of LSTM,the contextual information can be extracted by LSTM.Moreover,an attention mechanism is employed to assign weights to output from different time-series nodes within the LSTM,fo-cusing on key trajectory information.The experimental validation shows that the proposed model in com-parison with the LSTM model and the CNN-LSTM model,produces trajectory predictions to be even more close to match real trajectories.Specifically,the model in this paper achieves a 29.7%reduction in average prediction error compared to the LSTM model and a 25.4%reduction compared to the CNN-LSTM mod-el.In summary,the proposed method significantly enhances the accuracy of trajectory prediction.
作者
王堃
周志崇
曲凯
曹明松
胡延达
WANG Kun;ZHOU Zhichong;QU Kai;CAO Mingsong;HU Yanda(Unit 93886,Urumqi 830001,China;Air Traffic Control and Navigation School,Air Force Engineering University,Xi’an 710051,China;School of Computer Science,Shaanxi Normal University,Xi’an 710119,China)
出处
《空军工程大学学报》
CSCD
北大核心
2023年第6期50-57,共8页
Journal of Air Force Engineering University
关键词
航迹预测
注意力机制
卷积神经网络
循环神经网络
flight trajectory prediction
attention mechanism
convolutional neural network
recurrent neural network
作者简介
王堃(1986-),男,陕西汉中人,硕士,研究方向为空天防御作战指挥。E-mail:271969895@qq.com。
