Robotics simulation assets for
Unity
Build physics-enabled simulation environments for Unity. Rigyd converts 3D models, images, or text into assets with accurate mass, inertia tensors, friction, and collision geometry, ready for Unity Robotics Hub and ML-Agents.
The problem
Why existing workflows fall short.
Physics setup is per-GameObject
Every Rigidbody, Collider, and Physics Material has to be configured individually. Scaling from a few hero objects to a full training scene quickly becomes hours of manual component wiring.
Articulation Bodies need accurate inertia
Unity's Articulation system is only as realistic as its mass and inertia values. For custom assets, those numbers are usually guessed, and bad guesses break contact dynamics.
Asset pipeline is fragmented
FBX import, mesh cleanup, collider generation, physics material assignment, semantic tagging, every object runs through the same manual checklist. Diversity suffers.
How Rigyd helps
AI-native infrastructure that automates the hard parts.
Physics-ready FBX and OpenUSD output
Rigyd exports geometry with metadata for mass, collider type, and material physics so Unity's Rigidbody and Collider components populate automatically on import.
Calibrated inertia tensors
Volume analysis combined with material identification produces mass and inertia values within 15-20% of measured, ideal baselines for Articulation Body and ML-Agents training.
Bulk conversion for Robotics Hub
Enterprise API converts entire catalogs into Unity-ready assets in parallel, making it practical to build diverse, physically accurate training scenes for reinforcement learning.
cost reduction vs manual Unity physics setup
per asset, from upload to Unity-ready
mass accuracy vs measured values
Scale your Unity simulation environments
Drop in a 3D model, image, or text description and get Unity-ready, physics-enabled assets in minutes.
Frequently asked questions
How does Rigyd export assets to Unity?
Rigyd exports geometry as FBX or OpenUSD with metadata for mass, collider type, and physics materials. On import, Unity's Rigidbody and Collider components auto-populate, no manual wiring per GameObject.
Does Rigyd work with Unity ML-Agents?
Yes. Physics-enabled assets drop into ML-Agents scenes directly. Mass, inertia, and friction values are calibrated for contact-rich manipulation training, important for realistic RL policy transfer.
Can Rigyd populate Unity Articulation Bodies?
Yes. Rigyd estimates mass and inertia tensors per link with material-based accuracy. Articulation Body components inherit those values on import, avoiding the guesswork that breaks contact dynamics.
Does Rigyd work with Unity Robotics Hub out of the box?
Yes. For native USD workflows, Unity's USD Asset support reads Rigyd's OpenUSD directly. For Unity Robotics Hub's URDF Importer, the OpenUSD output can be converted to URDF using community USD→URDF tools and consumed directly. ROS-TCP-Connector integration works because the asset's physics, semantics, and joint metadata travel through both pipelines intact.
Can I use Rigyd assets with Unity URP or HDRP rendering pipelines?
Yes. Visual material bindings transfer to both URP and HDRP. PBR maps (base color, normal, roughness, metallic) populate Unity Shader Graph materials automatically. For HDRP, Rigyd preserves higher-fidelity material data like emissive maps and subsurface scattering parameters when present in the source file. Physics materials are tracked separately from visual materials, so rendering pipeline choice doesn't affect simulation accuracy.
Related reading
In-depth guides on robotics simulation, OpenUSD, and SimReady assets.
15 JUL 2026
Automatically Estimating Physics Properties for Simulation Assets
Mass, friction, and collision geometry are what make a 3D model behave in a physics engine. This is how those properties get estimated automatically, and why automating them is what lets teams build simulation assets at scale.
9 JUL 2026
Scaling Simulation Asset Libraries Beyond Curated Inventory
Curated SimReady libraries are a great starting point but a hard ceiling. Here is why fixed inventories limit robotics simulation at scale, and how on-demand asset generation closes the gap.
2 JUL 2026
Domain Randomization for Robotics Training: Asset Diversity at Scale
Domain randomization makes trained policies transfer to the real world, but it only works if your asset library is diverse enough. Here is how asset diversity drives randomization, and how to generate it at the scale training needs.
Explore more
SimReady for Isaac Sim
OpenUSD assets validated for NVIDIA Isaac Sim
Models for MuJoCo
Physically accurate assets for robot learning
Assets for Gazebo
Simulation-ready models for ROS 2
Warehouse Simulation
50,000+ SKUs need physics properties
Humanoid Robots
Whole-body interaction training assets
Sim-to-Real Transfer
Close the gap with accurate physics data