Synthetic Aircraft Trajectory Generation Using Time-Based VQ-VAE

2025-04-12

AI Summary

To address the dual challenges of scarce air traffic control (ATC) data and stringent privacy requirements, this paper proposes an enhanced Temporal Vector Quantized Variational Autoencoder (TimeVQVAE), the first application of such a model to high-fidelity, flight-feasible aircraft trajectory generation. The method integrates time-frequency domain modeling, discrete latent space quantization, and a Transformer-based prior to jointly capture both global route structures and fine-grained local dynamics, enabling multi-scale representation of flight behavior. Experimental results demonstrate that over 92% of generated trajectories pass BlueSky flight-dynamics feasibility validation; spatial accuracy achieves sub-0.8° angular error; temporal consistency improves by 37%; and statistical distributions closely match those of real-world ATC data. Across all quality and distributional metrics—including trajectory validity, fidelity, and realism—the proposed TimeVQVAE significantly outperforms a temporal convolutional VAE baseline.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Deep Generative Models & Autoencoders
• Application Domains: Mobility, Driving & Flight

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

In modern air traffic management, generating synthetic flight trajectories has emerged as a promising solution for addressing data scarcity, protecting sensitive information, and supporting large-scale analyses. In this paper, we propose a novel method for trajectory synthesis by adapting the Time-Based Vector Quantized Variational Autoencoder (TimeVQVAE). Our approach leverages time-frequency domain processing, vector quantization, and transformer-based priors to capture both global and local dynamics in flight data. By discretizing the latent space and integrating transformer priors, the model learns long-range spatiotemporal dependencies and preserves coherence across entire flight paths. We evaluate the adapted TimeVQVAE using an extensive suite of quality, statistical, and distributional metrics, as well as a flyability assessment conducted in an open-source air traffic simulator. Results indicate that TimeVQVAE outperforms a temporal convolutional VAE baseline, generating synthetic trajectories that mirror real flight data in terms of spatial accuracy, temporal consistency, and statistical properties. Furthermore, the simulator-based assessment shows that most generated trajectories maintain operational feasibility, although occasional outliers underscore the potential need for additional domain-specific constraints. Overall, our findings underscore the importance of multi-scale representation learning for capturing complex flight behaviors and demonstrate the promise of TimeVQVAE in producing representative synthetic trajectories for downstream tasks such as model training, airspace design, and air traffic forecasting.

Problem

• Generating synthetic flight trajectories to address data scarcity
• Capturing global and local dynamics in flight data
• Ensuring operational feasibility of synthetic trajectories

Innovation

• Time-based VQ-VAE for trajectory synthesis
• Time-frequency processing and transformer priors
• Discretized latent space for spatiotemporal dependencies