NVIDIA Isaac
The NVIDIA Isaac? AI robot development platform consists of NVIDIA? CUDA?-accelerated libraries, application frameworks, and AI models that accelerate the development of AI robots such as autonomous mobile robots (AMRs), arms and manipulators, and humanoids.
NVIDIA Isaac Libraries and AI Models
NVIDIA robotics full stack, CUDA-acceleration libraries, and optimized AI models give you a better, more efficient way to develop, train, simulate, deploy, operate, and optimize robot systems.
NVIDIA Isaac ROS
NVIDIA Isaac? ROS (Robot Operating System), built on the open-source ROS 2, is a collection of NVIDIA? CUDA?-accelerated computing packages and AI models designed to streamline and expedite the development of advanced AI robotics applications.
NVIDIA Isaac Manipulator
Isaac Manipulator, built on Isaac ROS, enables the development of AI-powered robotic arms that can seamlessly perceive, understand, and interact with their environments.
NVIDIA Isaac Perceptor
Isaac Perceptor, built on Isaac ROS, enables rapid development of advanced autonomous mobile robots (AMRs) that can perceive, localize, and operate in unstructured environments like warehouses or factories.
NVIDIA Isaac GR00T
NVIDIA Isaac GR00T is a research initiative and development platform for general-purpose robot foundation models and data pipelines to accelerate humanoid robotics.
Simulation and Robot Learning
Design, simulate, test, and train your AI-based robots and autonomous machines in a physically based virtual environment.
NVIDIA Isaac Sim
The NVIDIA Isaac Sim?, built on NVIDIA Omniverse?, helps you design, simulate, test, and train AI-based robots and autonomous machines in a physically based virtual environment.
NVIDIA Cosmos and Isaac Sim allow developers to generate synthetic data from 3D scenes for training perception robots.
NVIDIA Isaac Lab
This lightweight sample application is built on Isaac Sim and optimized for robot learning and robot foundation model training.
Develop, Test, and Optimize Physical AI and Robotic Fleets at Scale in Digital Twin Simulations
The Mega Omniverse Blueprint gives you a reference architecture of NVIDIA-accelerated computing, AI, Isaac, and Omniverse technologies to develop and test digital twins. You can use it with your own software for building AI-powered robot brains that drive robots, Metropolis perception from cameras, equipment, and more for continuous development, testing, and optimization.
Credit: KION Group, Accenture
NVIDIA-Accelerated Systems
NVIDIA’s three computing platforms streamline and accelerate developer workflows: NVIDIA DGX? systems for building robotics AI models, NVIDIA OVX? for simulating, testing, and training them, and NVIDIA AGX? for deploying and running them.
NVIDIA DGX
The NVIDIA DGX platform combines the best of NVIDIA software and infrastructure. It’s ideal for training multi-modal foundational models for robots.
NVIDIA OVX
NVIDIA OVX? systems provide industry-leading graphics and compute performance to accelerate the next generation of robotics.
NVIDIA AGX
NVIDIA AGX Systems, including NVIDIA Jetson?, offer high performance and energy efficiency, making them the leading platform for robotics. Trained, tested, and optimized robot AI models are deployed to these systems for real-world operation.
NVIDIA OSMO
NVIDIA OSMO is a cloud-native workflow orchestration platform that lets you easily scale your workloads across distributed environments—from on-premises to private and public cloud resource clusters.
