Resources
Field notes on building simulation assets.
Practical guides on physics estimation, SimReady structure, collision geometry, OpenUSD, and getting asset libraries to the scale robot training actually needs.
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.
Rigyd Team
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.
Rigyd Team
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.
Rigyd Team
Building a Scalable Embodied AI Asset Pipeline: From Raw Data to Simulation
A practical look at the stages of an embodied AI asset pipeline, from raw 3D data and reference inputs to physics-ready simulation assets, and what changes when you need to produce thousands of them instead of a handful.
Rigyd Team
Text and Image to Simulation Asset: Multi-Modal Input for Robotics Teams
How robotics and ML teams turn text descriptions, reference images, and existing 3D files into physics-ready simulation assets, and why multi-modal input matters when you are building training environments at scale.
Rigyd Team
What is a SimReady asset and why does it matter for robot learning and evaluation
A SimReady asset is a 3D model carrying the physics, collision, material, and semantic metadata a simulator needs to use it directly. This guide explains the anatomy of a SimReady asset and why it is the bottleneck for training and evaluating robot policies.
Rigyd Team
How Robotics Engineers and ML Teams Building Embodied AI Systems Can Convert 3D Models to Simulation Assets
A practical guide for robotics engineers and ML teams on converting 3D CAD models into physically accurate, simulation-ready assets for training and evaluating embodied AI systems at scale.
Rigyd Team
Rigyd vs NVIDIA SimReady Alternatives: A Guide for Robotics and ML Teams
Compare the best alternatives to NVIDIA SimReady asset libraries for robotics engineers and ML teams building physically accurate simulation environments.
Rigyd Team
Synthetic data generation for computer vision in robotics
Real-world labeled robotics data is expensive and slow to capture. Synthetic data is fast and unlimited, but only useful if the underlying simulation has correct physics, semantic labels, and domain randomization. Here's the complete pipeline.
Rigyd Team
Digital twin creation pipeline for manufacturing
A factory digital twin needs every object to behave physically, not just render. This is the end-to-end pipeline: CAD intake, BIM merge, physics layer, semantic labeling, simulation runtime, at the asset volumes (10K+ unique SKUs) real factories actually contain.
Rigyd Team
How to set up mass, friction, and joint properties for robot training
The three pillars of robot physics setup, mass, friction, joints, determine whether your trained policy transfers to real hardware. Here's the calibration target for each, the schemas, and the common mistakes that quietly break training.
Rigyd Team
Best practices for collision meshes in robotics sim
The collision mesh is the single most-tuned property in robotics simulation. Wrong choice = phantom interpenetration, slow physics, or both. Here's how to pick between primitives, convex hull, V-HACD decomposition, and mesh simplification, with concrete defaults per asset class.
Rigyd Team
How to convert GLTF/FBX/OBJ to OpenUSD for simulation
Most 3D pipelines export to GLTF, FBX, or OBJ. Most simulators expect OpenUSD. This guide covers the conversion paths, Omniverse, Blender, command-line, and AI-automated, plus what each format preserves and what gets lost.
Rigyd Team
Sim-to-real transfer: why physics accuracy matters more than visual fidelity
The sim-to-real gap is overwhelmingly a physics problem, not a rendering problem. Here's the research, the failure modes, the calibration ranges that actually matter, and why investing in physics accuracy beats investing in photorealism for most robotics policies.
Rigyd Team
Isaac Sim asset requirements and best practices
NVIDIA Isaac Sim expects assets in a specific format: OpenUSD with USDPhysics schemas, calibrated mass and inertia, convex collision meshes, semantic labels, and validated material bindings. This is the practical checklist.
Rigyd Team
How to create SimReady assets without manual modeling
Building a SimReady asset by hand takes ~4 hours and requires Blender, V-HACD, and USDPhysics expertise. AI-driven asset preparation collapses the workflow to about 5 minutes per asset. Here's how the automated pipeline works and when to use it.
Rigyd Team
What is OpenUSD and why does it matter for robotics
OpenUSD is the 3D scene description format Pixar built for film and NVIDIA scaled for robotics. This guide explains the architecture, the USDPhysics schemas, and why USD beats SDF/URDF/MJCF for modern simulation pipelines.
Rigyd Team
How to add physics properties to 3D models for simulation
Adding physics, mass, friction, collision meshes, inertia, is the bottleneck in robotics simulation. Here's the manual workflow, the AI-automated alternative, and a step-by-step guide for both.
Rigyd Team