Problem
OpenRadioss Car
Crash
A full vehicle in frontal impact, built from the open OpenRadioss Neon crash model and the Rescale AI Physics tutorial protocol. Surrogates predict how the body displaces over time as bumper, oil-pan, and pillar thicknesses change. Mesh resolution can differ by dataset. Transient crash response, not a static part study. Miss the fold pattern and the field score falls fast.
- Industry
- Automotive crash
- Domain
- Crash / FEA
- Task
- Transient displacement
- Datasets
- 2
Scope and limits
Comparability
Transient crash, not comparable with steady CFD or static structural problems. Mesh resolution differs by pack (decimated ~50k vs fuller Neon), so scores are fair within a dataset, not across packs. Public scoring uses median field R² across timesteps within each holdout case.
Ground truth
OpenRadioss exports built from the public Neon HPC benchmark model, following the Rescale AI Physics crash tutorial protocol. Holdout cases use simulation truth for nodal displacement. Today’s live scores are on the decimated surface extract.
Datasets
Same thickness DOE on the Neon frontal-crash story. The live pack is the Rescale Tutorial on a decimated ~50k-node surface extract; a fuller Neon mesh pack is listed for later.
1 live · 2 total
| Dataset | What varies | Geometry | Mesh | Solver | Cases | Runs | Open |
|---|---|---|---|---|---|---|---|
Rescale Tutorial Rescale · Live | Four shell thicknesses | Neon ~50k | Decimated | OpenRadioss | 40cases | 2 runs | Open |
Neon Full OpenRadioss · Soon | Four shell thicknesses | Neon ~1M | Full | OpenRadioss | 40cases | — | Soon |