AB-UPT: Scaling Neural CFD Surrogates for High-Fidelity Automotive Aerodynamics Simulations via Anchored-Branched Universal Physics Transformers
Paper
•
2502.09692
•
Published
•
1
Emmi AI developed and released Anchor-Branched Universal Physics Transformers (AB-UPT) model for aerodynamics simulations. The model has been trained on DrivAerML dataset.
It is expected that the user will provide input geometry similar to the ones from the DrivAerML dataset.
We provide a standalone GitHub repository with relevant tutorial code on how execute inference. The corresponding repository can be found here: https://github.com/Emmi-AI/AB-UPT
AB-UPT is shared under the CC-BY-NC 4.0 license.
@article{alkin2025abupt,
title={{AB-UPT}: Scaling Neural CFD Surrogates for High-Fidelity Automotive Aerodynamics Simulations via Anchored-Branched Universal Physics Transformers},
author={Benedikt Alkin and Maurits Bleeker and Richard Kurle and Tobias Kronlachner and Reinhard Sonnleitner and Matthias Dorfer and Johannes Brandstetter},
journal={arXiv preprint arXiv:2502.09692},
year={2025}
}