Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DA-STGCN: 4D Trajectory Prediction Based on Spatiotemporal Feature Extraction

Yuheng Kuang, Zhengning Wang, Jianping Zhang, Zhenyu Shi, Yuding Zhang

TL;DR

DA-STGCN is proposed, an innovative spatiotemporal graph convolutional network that integrates a dual attention mechanism that reconstructs the adjacency matrix through a self-attention approach, enhancing the capture of node correlations, and employs graph attention to distill spatiotemporal characteristics, thereby generating a probabilistic distribution of predicted trajectories.

Abstract

The importance of four-dimensional (4D) trajectory prediction within air traffic management systems is on the rise. Key operations such as conflict detection and resolution, aircraft anomaly monitoring, and the management of congested flight paths are increasingly reliant on this foundational technology, underscoring the urgent demand for intelligent solutions. The dynamics in airport terminal zones and crowded airspaces are intricate and ever-changing; however, current methodologies do not sufficiently account for the interactions among aircraft. To tackle these challenges, we propose DA-STGCN, an innovative spatiotemporal graph convolutional network that integrates a dual attention mechanism. Our model reconstructs the adjacency matrix through a self-attention approach, enhancing the capture of node correlations, and employs graph attention to distill spatiotemporal characteristics, thereby generating a probabilistic distribution of predicted trajectories. This novel adjacency matrix, reconstructed with the self-attention mechanism, is dynamically optimized throughout the network's training process, offering a more nuanced reflection of the inter-node relationships compared to traditional algorithms. The performance of the model is validated on two ADS-B datasets, one near the airport terminal area and the other in dense airspace. Experimental results demonstrate a notable improvement over current 4D trajectory prediction methods, achieving a 20% and 30% reduction in the Average Displacement Error (ADE) and Final Displacement Error (FDE), respectively. The incorporation of a Dual-Attention module has been shown to significantly enhance the extraction of node correlations, as verified by ablation experiments.

DA-STGCN: 4D Trajectory Prediction Based on Spatiotemporal Feature Extraction

TL;DR

DA-STGCN is proposed, an innovative spatiotemporal graph convolutional network that integrates a dual attention mechanism that reconstructs the adjacency matrix through a self-attention approach, enhancing the capture of node correlations, and employs graph attention to distill spatiotemporal characteristics, thereby generating a probabilistic distribution of predicted trajectories.

Abstract

The importance of four-dimensional (4D) trajectory prediction within air traffic management systems is on the rise. Key operations such as conflict detection and resolution, aircraft anomaly monitoring, and the management of congested flight paths are increasingly reliant on this foundational technology, underscoring the urgent demand for intelligent solutions. The dynamics in airport terminal zones and crowded airspaces are intricate and ever-changing; however, current methodologies do not sufficiently account for the interactions among aircraft. To tackle these challenges, we propose DA-STGCN, an innovative spatiotemporal graph convolutional network that integrates a dual attention mechanism. Our model reconstructs the adjacency matrix through a self-attention approach, enhancing the capture of node correlations, and employs graph attention to distill spatiotemporal characteristics, thereby generating a probabilistic distribution of predicted trajectories. This novel adjacency matrix, reconstructed with the self-attention mechanism, is dynamically optimized throughout the network's training process, offering a more nuanced reflection of the inter-node relationships compared to traditional algorithms. The performance of the model is validated on two ADS-B datasets, one near the airport terminal area and the other in dense airspace. Experimental results demonstrate a notable improvement over current 4D trajectory prediction methods, achieving a 20% and 30% reduction in the Average Displacement Error (ADE) and Final Displacement Error (FDE), respectively. The incorporation of a Dual-Attention module has been shown to significantly enhance the extraction of node correlations, as verified by ablation experiments.

Paper Structure

This paper contains 14 sections, 14 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related works
  3. Methodology
  4. The DA-STGCN model
  5. Construction of spatiotemporal graph
  6. Reconstruct adjacency matrix based on attention
  7. Feature extraction by STGCN and GAT
  8. Trajectory prediction and loss design
  9. Experiment and Evaluation
  10. Datasets and evaluation metrics
  11. Experiment settings
  12. Analysis of experimental results
  13. Conclusion
  14. Acknowledgments

Figures (6)

  • Figure 1: The actual flight path of aircraft sometimes doesn't follow the pre-planned route. Advanced 4D trajectory prediction can assist ATM in better handling these emergencies.
  • Figure 2: Overall framework of DA-STGCN model.
  • Figure 3: Adjacency Matrix Reconstruction Module Based on Self-Attention Mechanism.
  • Figure 4: Operation STGCN and GAT.
  • Figure 5: Qualitative Analysis of Aircraft Trajectory Prediction Results. (a) and (b) show the trajectory distributions in the horizontal and vertical dimensions predicted by the DA-STGCN model for a scenario in the airport terminal area, while (c) and (d) display the corresponding predictions by the Social-STGCNN model.
  • ...and 1 more figures