Physical AI-powered robots need to autonomously sense, plan, and perform complex tasks. Robot simulation allows developers to train, simulate, and validate these systems in virtual environments prior to real-world deployment. This guide explains how to use ROS 2 with NVIDIA Isaac Sim to create and test robotic workflows, including adding sensors, connecting ROS 2 actions, and using synthetic data for model training. Isaac Sim offers a scalable and extensible platform built on OpenUSD, supporting diverse simulations from simple office environments to complex factories.
Table of contents
Isaac Sim with ROS 2 workflowURDF: A common starting point for simulation in Isaac SimPrepopulated scenes and SimReady assetsAdding sensorsConnecting to ROS 2Moving to perception AI-enabled robotsMore workflows and use casesGet started with robotic simulationSort: