Ground-based Wind Field Construction from Mode-S and ADS-B Data with a Novel Gas Particle Model

Ground-based Wind Field Construction from Mode-S and ADS-B Data with a Novel Gas Particle Model

Sun, Junzi (TU Delft Control & Simulation)
Vû, Huy
Ellerbroek, Joost (TU Delft Control & Simulation)
Hoekstra, J.M. (TU Delft Control & Simulation)

2017

Wind is an important parameter in many air traffic management researches, as it often introduces significant uncertainties in aircraft performance studies and trajectory predictions. Obtaining accurate wind field information has always been a challenge due to the availability of weather sensors. Traditionally, there is no direct method to measure wind data at different altitudes with the exception of weather balloon systems that cannot be easily scaled. On the other hand, aircraft, which rely heavily on atmospheric data, can be part of atmospheric model itself. Aircraft can provide wind and temperature measurements to ground observers. In this paper, aircraft are considered as a moving sensor network established to re-construct the wind field on a larger scale. Based on the powerful open-source tool pyModeS, aircraft ground velocity and airspeed are decoded from ADS-B and Mode-S data respectively. Wind observations are then derived based on the difference of these two vectors. An innovative gas particle model is also developed so that the complete wind field can be constructed continuously based on these observations. The model can generate wind field in real-time and at all flight levels. Furthermore, the confidence of wind at any 4D position can be computed according to the proposed model method. Multiple selfand cross-validations are conducted to ensure the correctness and stability of the model, as well as the resulting wind field. This paper provides a series of novel methods, as well as open-source tools, that enable the research community using simple ADS-B/Mode-S receivers to construct accurate wind fields.

ADS-B
Mode-S
aviation weather
wind modeling
aircraft sensor network
gas particle model

http://resolver.tudelft.nl/uuid:cf180450-cf66-4532-9c0d-ff64159b0901

Proceedings of the Seventh SESAR Innovation Days, 28th – 30th November 2017: Belgrade, Serbia

7th SESAR Innovation Days, 2017-11-28 → 2017-11-30, University of Belgrade, Belgrade, Serbia

Institutional Repository

conference paper

© 2017 Junzi Sun, Huy Vû, Joost Ellerbroek, J.M. Hoekstra