Learning Generative Models for Climbing Aircraft from Radar Data

TL;DR

This paper introduces a data-driven generative model that enhances aircraft trajectory predictions during climb phases by correcting BADA model outputs, resulting in more accurate and realistic forecasts with confidence bounds.

Contribution

It presents a novel generative approach that learns functional corrections to the BADA model from radar data, improving trajectory prediction accuracy and providing confidence bounds.

Findings

26.7% reduction in arrival time prediction error

Generated trajectories closely match test data

Efficient computation of confidence bounds

Abstract

Accurate trajectory prediction (TP) for climbing aircraft is hampered by the presence of epistemic uncertainties concerning aircraft operation, which can lead to significant misspecification between predicted and observed trajectories. This paper proposes a generative model for climbing aircraft in which the standard Base of Aircraft Data (BADA) model is enriched by a functional correction to the thrust that is learned from data. The method offers three features: predictions of the arrival time with 26.7% less error when compared to BADA; generated trajectories that are realistic when compared to test data; and a means of computing confidence bounds for minimal computational cost.