PhysicsX tests car-aerodynamics surrogates on new vehicles with its 20,000-run PXNetCar dataset

PhysicsX argues that DrivAerNet++ (8,121 geometries from three baselines) and Luminary SHIFT-SUV (1,994 geometries from two baselines) vary too little to test generalisation, because random splits put morphs of the same baseline in both training and test sets. It built PXNetCar, a proprietary dataset of 20,486 RANS simulations in STAR-CCM+ over 260 baseline vehicles with about 100 variants each. PhysicsX reports that state-of-the-art architectures generalise poorly to new vehicles and across CFD setups, that out-of-distribution error falls steadily as training data becomes more diverse, and that training on several datasets recovers in-simulator accuracy. The dataset is not released.

Share on LinkedIn, X or Bluesky, or copy the link.