SCENEGRAPH
USD is fundamentally about scalable scene description and composition.
OpenUSD was created to robustly interchange and augment complex 3D scenes assembled from many assets. Its composition model supports references, variants and sparse overrides while presenting a unified hierarchical scenegraph.
FILE ENCODINGS
USD is a system with multiple native encodings.
.usda
Human-readable UTF-8 text representation.
.usdc
Binary Crate representation designed for efficient access.
.usd
Generic extension that can contain either native representation.
References
Assets can be composed from other layers instead of flattened into one file.
Variants
Different configurations can coexist as composable scene opinions.
Time-sampled data
Scene description can include time-varying values.
USD PHYSICS
Robotics gains rigid bodies, collisions and joints without turning USD into a robot-only format.
UsdPhysics provides schemas for rigid bodies, colliders, physical materials, joints and simulation scenes. That makes USD particularly useful in rich simulation, synthetic-data and digital-twin workflows.
OPENRGD BOUNDARY
OpenRGD does not pretend its lightweight parser is a full USD engine.
The current OpenRGD USDA importer intentionally supports a narrow auditable subset. It does not compose layers, resolve references or payloads, evaluate variants or replace the OpenUSD SDK. A source requiring those semantics should use a future full OpenUSD adapter.
COMPARE
USD and RGD are complementary when scene context meets semantic embodiment.
Links, joints, geometry, inertials, collision and joint constraints in a ROS-centric XML model.
Not a general scene-composition system or semantic embodiment graph.
USDComposable scene descriptionLarge hierarchical scenegraphs, references, variants, time-sampled data and physics schemas.
Not a robot-specific semantic capability or governance model by default.
SDFSimulation-oriented model and world descriptionRobots, environments, models, sensors, physics, lights, plugins and simulation context.
Not designed as a general AI embodiment-governance or provenance graph.
MJCFMuJoCo-native dynamic-system modelNested bodies, joints, actuators, tendons, sensors, defaults and simulator-native dynamic modeling.
Optimized for MuJoCo rather than cross-stack semantic embodiment.
RGDSemantic embodiment graphIdentity, physical evidence, operational constraints, capability semantics, provenance and lifecycle context.
Does not replace scene composition, a physics engine, middleware or a physical runtime.