Deep generative modelling of aircraft trajectories in terminal maneuvering areas

• Published in: Machine learning with applications Vol. 11; p. 100446
• Main Authors: Krauth, Timothé, Lafage, Adrien, Morio, Jérôme, Olive, Xavier, Waltert, Manuel
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.03.2023; Elsevier
• Subjects: Air traffic management; Deep Generative Models; Engineering Sciences; Mathematics; Multivariate time-series generation; Physics; Temporal Convolutional Networks; Variational Autoencoder; Deep Generative Models; Multivariate time-series generation; Variational Autoencoder; Air traffic management; Temporal Convolutional Networks; Multivariate Time-Series Generation; Air Traffic Management
• ISSN: 2666-8270; 2666-8270
• DOI: 10.1016/j.mlwa.2022.100446

Airspace design is subject to a multitude of constraints, which are mainly driven by the concern to keep the risk of mid-air collision below a target level of safety. For that purpose, Monte Carlo simulation methods can be applied to estimate aircraft conflict probability but require the accurate generation of artificial trajectories. Generative models allow to generate an infinite number of trajectories for air traffic procedures where only few observations are available. The generated trajectories must not only resemble observed trajectories in terms of statistical distributions but they should stay flyable and consider uncertainty due to weather, air traffic control, aircraft performances, or human factors. This paper focuses on the generation problem, and its main contribution lies in the adaptation of the Variational Autoencoder structure to the problem of 4-dimensional aircraft trajectories modelling using Temporal Convolutional Networks and a prior distribution composed of a Variational Mixture of Posteriors (VampPrior). The proposed model has been trained on trajectories in the Terminal Manoeuvre Area of Zurich airport, which have a particularly high degree of variability as air traffic controllers often take actions that deviate aircraft from the nominal approach procedure. The model has demonstrated great abilities to take into account such amount of uncertainty. Regarding metrics that evaluate the estimation of the statistical distribution of the observed trajectories, and the flyability of the generated ones, the proposed method outperforms traditional statistical methods by being able to generate more complex and realistic trajectories. •Collision risk estimation requires a large number of diverse aircraft trajectories.•Variational Autoencoders (VAE) can generate diverse but realistic trajectories.•The VAE can be trained with a focus on the most recent flight procedures.•The VAE generation process preserves the distribution of observed trajectories.•The VAE generation process allows creating high risks trajectories.