This paper introduces a novel prediction model designed to mitigate the substantial data dependency associated with maneuver trajectory prediction in unmanned combat air vehicles (UCAVs) during air combat. Considering the characteristics of high noise, dynamic complexity, and variable data lengths inherent in short-range air combat scenarios, we employ dynamic time warping (DTW) to assess the similarity of 3D trajectory data. This approach allows us to identify and select the most analogous historical data, which we then utilize as our training dataset. In pursuit of enhanced precision for online trajectory prediction, we propose an improved convolutional neural network (CNN) that not only offers “after-zero” information but also incorporates delay compensation mechanisms. Our experimental findings indicate that the proposed prediction model not only satisfies the stringent timeliness requirements but also outperforms benchmark models in terms of prediction accuracy across various operating conditions.
Short-Term Target Maneuvering Trajectory Prediction Using DTW–CNN–LSTM
2025
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Short-Term Traffic Flow Prediction: Using LSTM
IEEE | 2020
|STG-LSTM: Spatial-temporal graph-based long short-term memory for vehicle trajectory prediction
DOAJ | 2025
|Short-term trajectory prediction methods
AIAA | 1999
|Short-Term Trajectory Prediction Methods
British Library Conference Proceedings | 1999
|