Physics-informed machine learning for wind turbulence reconstruction with lidar under aircraft motion

Amaury Capmas-Pernet, Christian Musso, Frédéric Dambreville, Tomline Michel
Proceedings of the 2nd International Conference on Probabilistic Numerics, PMLR 341:1-8, 2026.

Abstract

We propose a physics-informed machine learning method for reconstructing 3d turbulent wind from partial, noisy and accumulated over time observations obtained via lidar measurements. This method, which is optimal in terms of mean squared error, combines Gaussian process regression with a physics-informed kernel that leverages prior knowledge of the turbulence structure, as provided by the von Karman turbulence model.

Cite this Paper

BibTeX
@InProceedings{pmlr-v341-capmas-pernet26a, title = {Physics-informed machine learning for wind turbulence reconstruction with lidar under aircraft motion}, author = {Capmas-Pernet, Amaury and Musso, Christian and Dambreville, Fr\'ed\'eric and Michel, Tomline}, booktitle = {Proceedings of the 2nd International Conference on Probabilistic Numerics}, pages = {1--8}, year = {2026}, editor = {Karvonen, Toni and Bosch, Nathanael and Cockayne, Jon and Gessner, Alexandra and Hennig, Philipp and Kouw, Wouter}, volume = {341}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v341/main/assets/capmas-pernet26a/capmas-pernet26a.pdf}, url = {https://proceedings.mlr.press/v341/capmas-pernet26a.html}, abstract = {We propose a physics-informed machine learning method for reconstructing 3d turbulent wind from partial, noisy and accumulated over time observations obtained via lidar measurements. This method, which is optimal in terms of mean squared error, combines Gaussian process regression with a physics-informed kernel that leverages prior knowledge of the turbulence structure, as provided by the von Karman turbulence model.} }
Endnote
%0 Conference Paper %T Physics-informed machine learning for wind turbulence reconstruction with lidar under aircraft motion %A Amaury Capmas-Pernet %A Christian Musso %A Frédéric Dambreville %A Tomline Michel %B Proceedings of the 2nd International Conference on Probabilistic Numerics %C Proceedings of Machine Learning Research %D 2026 %E Toni Karvonen %E Nathanael Bosch %E Jon Cockayne %E Alexandra Gessner %E Philipp Hennig %E Wouter Kouw %F pmlr-v341-capmas-pernet26a %I PMLR %P 1--8 %U https://proceedings.mlr.press/v341/capmas-pernet26a.html %V 341 %X We propose a physics-informed machine learning method for reconstructing 3d turbulent wind from partial, noisy and accumulated over time observations obtained via lidar measurements. This method, which is optimal in terms of mean squared error, combines Gaussian process regression with a physics-informed kernel that leverages prior knowledge of the turbulence structure, as provided by the von Karman turbulence model.
APA
Capmas-Pernet, A., Musso, C., Dambreville, F. & Michel, T.. (2026). Physics-informed machine learning for wind turbulence reconstruction with lidar under aircraft motion. Proceedings of the 2nd International Conference on Probabilistic Numerics, in Proceedings of Machine Learning Research 341:1-8 Available from https://proceedings.mlr.press/v341/capmas-pernet26a.html.

Related Material