Home // International Journal On Advances in Intelligent Systems, volume 11, numbers 3 and 4, 2018 // View article

Air Traffic Representation and Analysis Through Local Covariance

Authors:
Georges Mykoniatis
Florence Nicol
Stephane Puechmorel

Keywords: Air traffic complexity; spatial data; manifold valued images; covariance function estimation; non-parametric estimation

Abstract:
Air traffic is generally characterized by simple indicators like the number of aircraft flying over a given area or the total distance flown during a time window. As an example, these values may be used for estimating a rough number of air traffic controllers needed in a given control center or for performing economic studies. However, this approach is not adapted to more complex situations such as those encountered in airspace comparison or air traffic controllers training or for adapting dynamically the airspace configurations to the traffic conditions. An innovative representation of the traffic data, relying on a sound theoretical framework, is introduced in this work. It will pave the way to a number of tools dedicated to traffic analysis. Based on an extraction of local covariance, a grid with values in the space of symmetric positive definite matrices is obtained. It can serve as a basis of comparison or be subject to filtering and selection to obtain a digest of a traffic situation suitable for efficient complexity assessment.

Pages: 268 to 278

Copyright: Copyright (c) to authors, 2018. Used with permission.

Publication date: December 30, 2018

Published in: journal

ISSN: 1942-2679