Hugging Face Launches SimReady Central Library for Robotics and AI Factory Simulations

The library includes infrastructure and equipment assets such as robot arms, grippers, warehouse equipment, racks and cooling infrastructure. Developers can check which capabilities each asset supports and combine assets from different suppliers into a scene, according to the post.

SimReady assets are OpenUSD assets prepared with the properties required for simulation workflows, the post says. A detailed 3D model such as CAD is a starting point, but simulation needs additional information about how the object should be interpreted and used. For robot manipulation, that can include mass, collision geometry and friction properties; for AI factory workflows, it can include equipment geometry and the electrical or thermal metadata required by the application.

OpenUSD is an open, extensible framework for describing and composing 3D scenes. Using those scenes to train and evaluate autonomous machines also requires information about physical behavior, collisions and sensor interactions, according to the post. SimReady Foundation provides specifications, profiles and validation tools that make these requirements explicit; a profile defines the logical grouping of features and capabilities an asset needs for a particular use case.

Depending on the asset and profile, features can include collision geometry and physical properties for contact and physics simulation; joint structure and limits for articulated robot and equipment movement; semantic labels for object identification and annotated synthetic data; and materials and sensor attributes for rendering and sensor simulation. Developers should select assets whose profiles match their workflow needs and compose them into an OpenUSD scene, the post says.

Equipment manufacturers can prepare and validate assets against SimReady requirements and use SimReady Central Library to reach developers, according to the post. The library is intended to let end customers test and simulate scenarios faster and make decisions sooner, and to let industries build simulations around their products while managing their own Hugging Face datasets. OEMs can turn CAD equipment models into SimReady assets for developers building simulations around their products.

The post gives examples across equipment and robotics. For AI factory OEMs such as Vertiv, the library offers a path to distribute assets including warehouse equipment racks, server racks, cooling equipment and supporting infrastructure for AI factory digital twins. For robotics OEMs including Comau, FANUC, Franka Robotics, Robotiq, Standard Bots and Universal Robots, it makes robot arms and grippers discoverable to teams building training and evaluation environments; developers can combine these with industrial equipment to build warehouse and factory scenarios.

For simulation asset provider Palatial, the library features the NVIDIA Robotics 1K Dataset released in collaboration with NVIDIA Robotics, along with complex articulated assets generated from text, images, scans, datasheets and CAD. For LG Electronics, which develops home robots alongside appliances, the library provides a way to share assets such as refrigerators and dishwashers; developers can use assets informed by the company's experience in appliances and robotics to build household simulation environments.