High-Fidelity Road-Car Aerodynamics CFD Dataset

DrivAerML provides a dataset for road-car external aerodynamics generated using high-fidelity Computational Fluid Dynamics simulations. It was created by Neil Ashton to address the lack of open-source training data for realistic automotive shapes. The dataset was last updated on the platform in May 2025.

Use Cases

• Predict pressure coefficients or drag forces from vehicle geometry and flow conditions using CFD simulation features.
• Train generative models to design new car shapes that optimize aerodynamic performance metrics.
• Develop reduced-order models for real-time flow field prediction around road cars.
• Benchmark machine learning algorithms on high-dimensional physics simulation outputs.

Strengths

• Dataset focuses on high-fidelity CFD simulations for realistic road-car geometries.
• Created specifically to address a known gap in open-source automotive aerodynamics data.

Limitations

• Specific dataset size, row count, and column details are not provided in the input.
• The full description and data specifics are hosted externally, requiring a page visit.

Provenance

Source

https://caemldatasets.org

Collection Method

Generated via high-fidelity Computational Fluid Dynamics simulations.

Freshness

Last updated on the platform on 2025-05-23.