Trajectory prediction method for incoming guided projectiles based on the fusion of Temporal Convolutional Network and dual attention mechanisms

To improve the response speed and missile interception accuracy of combat systems, this paper proposes a trajectory prediction method for incoming projectiles based on the TCN-ECA-BiGRU-Attention model. First, trajectory prediction datasets are constructed under various combat scenarios. Compared wi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Measurement : journal of the International Measurement Confederation Jg. 257; S. 118906
Hauptverfasser: Gao, Zhanpeng, Yi, Wenjun
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 15.01.2026
Schlagworte:
Anti-missile interception
Attention mechanism
Bidirectional gated recurrent unit
Temporal convolutional network
Trajectory prediction
Attention mechanism
Trajectory prediction
Temporal convolutional network
Bidirectional gated recurrent unit
Anti-missile interception
ISSN:0263-2241
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To improve the response speed and missile interception accuracy of combat systems, this paper proposes a trajectory prediction method for incoming projectiles based on the TCN-ECA-BiGRU-Attention model. First, trajectory prediction datasets are constructed under various combat scenarios. Compared with BiGRU, BiLSTM, CNN-BiGRU, CNN-BiGRU-Attention, and TCN-BiGRU-Attention models, simulation results show that the TCN-ECA-BiGRU-Attention model achieves the highest prediction accuracy across different operational environments. For guided artillery targeting stationary points, the cumulative error over 10 s in the three axes is only about 60 meters. For guided artillery targeting moving points, the overall prediction error is approximately 200 meters, which falls within the lethal range of interception missiles. Moreover, the model’s average trajectory point error is smaller than its endpoint error, demonstrating high stability and accuracy. Compared with other models, the TCN-ECA-BiGRU-Attention model exhibits lower endpoint and average trajectory errors, highlighting its superior precision and reliability. Therefore, the proposed trajectory prediction method not only effectively captures the dynamic behavior of targets but also provides reliable interception data for guidance systems, thereby enhancing operational efficiency in modern warfare. •Designs guidance law to simulate artillery trajectory data for TEBGA training and validation.•TEBGA hybrid with recursive prediction enables accurate trajectory forecasting.•Validates TEBGA’s superior precision, stability and generalization against traditional models.
ISSN:0263-2241
DOI:10.1016/j.measurement.2025.118906