Validate a Composed SimReady Scene#

This page expands the answer to How do I validate my composed scene of SimReady assets to run my simulation?.

Say I want to run a particular simulation, such as electrical, robotics, or autonomous vehicle, and I need to ensure all the SimReady assets in my composed scene are connected appropriately, structurally complete, and collectively ready for the target runtime. How do you validate the scene before running it?

Start with assets that conform to the profiles required by your workflow, reusing existing conformance results where they remain applicable. Verify the composed scene’s references, units, connections and configuration against the target workflow’s requirements, then evaluate its behavior in the target runtime against expected results. Changes or overrides that affect validated properties may require affected checks to run again. Asset conformance, scene integration and runtime evaluation provide complementary evidence. For asset conformance detail, refer to Validate SimReady Assets with Schemas.

Asset, Scene and Runtime Checks#

These checks cover three scopes:

  • Asset conformance: Confirm that each asset conforms to the profiles the workflow requires. Reuse an existing conformance result when it still applies to the asset and profile version. A reported failure traces to a specific requirement and USD property.

  • Scene integration: Verify the composed scene’s references, units, connections, and configuration against the target workflow’s requirements.

  • Runtime evaluation: Load the composed scene in the target runtime, such as Omniverse Isaac Sim, Isaac Lab, MEGA, AVSim, or DRIVE Sim, and compare its behavior with the expected results.

Changes or overrides that affect validated properties may require the affected checks to run again.

Static Scene-Level Validation#

Static scene validation is not behavior testing. It examines the composed USD stage for structure, relationships, and declared intent without running the simulation. The distinction is the same as compiling code and executing it: structural checks can catch mismatched types, missing references, and unresolved symbols without running the simulation.

These scene checks establish the following:

  • MEGA scenes: The scene contains a discoverable environment, robots, and sensor attachments in the expected hierarchy, and the metadata required to derive the MEGA runtime configuration is present.

  • AVSim scenes: The scene includes the required vehicles, traffic elements, and map structure, and the runtime configuration can be derived from scene metadata.

  • Sensor attachment: Sensors are structurally attached to robots or vehicles and are not orphaned prims in the scene graph.

  • Cross-asset references: References resolve, composition arcs are intact, and the composition patterns are compatible across assets that were individually valid.

  • Runtime configuration derivability: The scene’s declared metadata is sufficient for the target simulator to construct its runtime configuration.

These checks are structural and declarative, not behavioral. They establish whether the scene is well-formed for a workflow. They do not establish that the scene produces the expected simulation results.

Selecting Checks for Your Workflow#

Available checks and supported workflows are determined by the applicable versioned specifications and tool documentation. Structural checking examines authored structure, relationships, and declared intent. Behavioral evaluation runs the scene in the target runtime and compares the results with what the workflow expects. The two kinds of evidence are complementary. Static analysis does not establish that a physically simulated scene behaves correctly.