Progressive multi-resolution training: A training recipe for fast, three-dimensional high-resolution aerodynamic prediction

The aerodynamic optimization of cars requires close collaboration between aerodynamicists and stylists, while slow, expensive simulations remain a bottleneck. Surrogate models have been shown to accurately predict aerodynamics within the design space for which they were trained. However, many of these models struggle to scale to higher resolutions because of the three-dimensional nature of the problem and data scarcity. We propose progressive multi-resolution training (PMRT), a probabilistic multi-resolution training schedule that enables training a U-Net to predict the drag coefficient (cd) and high-resolution velocity fields (512 × 128 × 128) in 24 h on a single NVIDIA H100 graphics processing unit, 7× cheaper than the high-resolution-only baseline, with similar accuracy. PMRT samples batches from three resolutions based on probabilities that change during training, starting with an emphasis on lower resolutions and gradually shifting toward higher resolutions. Since this is a training methodology, it can be adapted to other high-resolution-focused backbones. We also show that a single model can be trained across five datasets from different solvers, including a real-world dataset, by conditioning on the simulation parameters. We train both cd-only and cd + velocity models to cover applications where flow fields are either unavailable or unnecessary. In the DrivAerML dataset, our models achieve a cdR2 value of ≈0.974, matching literature baselines at a fraction of the training cost. The reduced training time keeps the time-to-first prediction on new data short, which is essential for real-world iterative vehicle development.