When simulating real-world systems in Isaac Sim—like factories, warehouses, or robotics testbeds—it’s essential to think in terms of processes. A process-oriented simulation models how each entity moves through a sequence of tasks over time.

Build discrete-event simulation logic on top of physics-enabled USD scenes to model, visualize, and analyze complex job-shop workflows with real-time state transitions, resource contention, and dynamic task routing in a fully programmable environment.

From flowcharts to physics-based scenes, simulation has evolved—and Champion is leading the shift with USD and AI.

A side-by-side mapping of classical simlib simulation functions to their modern equivalents in Champion’s agentic AI system, showing how AI automates scene logic, scheduling, and physics using USD and Python.

Behind every reliable simulation is a structured scene. USD is the format that makes it all work.

It’s any scenario where multiple agents (robots, vehicles, etc.) arrive over time and need access to a limited number of resources (like docks, tools, or processors).

Even the most advanced simulation fails without sound modeling, clean data, and the right engineering process—here’s how to avoid the most common pitfalls.

Most teams waste days fixing CAD files, debugging physics, or rewriting scripts. It's time to automate the pipeline and scale your simulation efforts.

Before you can simulate anything, you need to define the system you're simulating. In NVIDIA Omniverse, that system takes shape as a USD stage—a powerful, layered scene graph that holds your geometry, physics settings, and custom attributes.