The model will be trained on the world's largest open-source aerodynamics database of 25,000 simulations.
Luminary Cloud, the leading scalable engineering simulation platform, announced SHIFT Models, its suite of foundation models. The initial SHIFT-SUV model is the world’s first Physics AI foundation model and dataset designed for industrially-relevant automotive SUV aerodynamic analysis. Developed in collaboration with Honda and NVIDIA, this groundbreaking model will enable automotive designers and engineers to perform real-time interactive aerodynamic analysis during the initial design process.
A critical need in automotive aerodynamics development
This foundation model, combined with Luminary’s Physics AI virtual wind tunnel, addresses a critical need in the automotive industry by bridging the gap between design aesthetics and engineering performance. Currently designers and engineers are inhibited by the inability to quickly make performance-based decisions in early stage development. While there are a few open-source simulation databases for sedans based on the DrivAer platform, automakers lack a strong foundation model and simulation database for SUVs—the fastest-growing automotive segment globally, accounting for 48% of global car sales in 2023.
A new modern Physics AI approach to automotive design
The SHIFT-SUV aerodynamic dataset was generated by simulating several thousand geometry variants of the AeroSUV open-geometry model with Luminary’s high-fidelity Computational Fluid Dynamics (CFD) platform. The foundational model is trained on this dataset using NVIDIA’s PhysicsNeMo, leveraging its DoMINO architecture for automotive external-aerodynamics, enabling Physics AI-based simulation (inference). The AI-based workflow is superior to traditional design approaches as it avoids cumbersome workflow steps, providing designers with instant and accurate aerodynamic performance insights. Luminary Cloud’s model is deployed in a virtual wind tunnel, allowing designers and engineers to test designs in real-world simulations to bridge the gap between design aesthetics and engineering performance.
At launch, the SHIFT-SUV model is trained on approximately one thousand simulations, with a goal of advancing to 25,000 by the end of the year. Luminary Cloud will rollout new data monthly, with a staged release of pre-trained models at specific dataset size milestones. The dataset within Luminary Cloud’s model will be open-source.
The foundation model offers several key benefits for the automotive industry, such as:
- Access to difficult to find (or generate) open-source, industry-relevant data. Luminary Cloud will release both the dataset and foundation model, enabling widespread adoption and customization.
- The largest open-source Physics AI SUV aerodynamics foundation model database. The high-fidelity, transient DDES CFD simulations include various vehicle configurations, body types, and flow conditions–increasing accuracy in the AI model prediction.
- Contribution from Honda to generate design variations to ensure industrially-relevant geometry training set, along with NVIDIA’s accelerated GPU computing and NVIDIA Omniverse Blueprint for real-time digital twins CAE resources to accelerate simulation database generation and model training with NVIDIA PhysicsNeMo.
Available now, automotive OEMs, suppliers, and research institutions that are interested in the SHIFT-SUV foundation model and dataset can access it by contacting shift.luminarycloud.com.
For more information, visit luminarycloud.com.
