NVIDIA Fuels the Physical AI Revolution with Innovative Robotics Tools
NVIDIA is celebrating significant advancements in physical AI during National Robotics Week 2026. They are showcasing new technologies, unveiled at GTC last month, which are propelling the growth of AI-powered robots across various sectors.
At the heart of NVIDIA’s robotics initiative is a comprehensive, cloud-to-robot workflow. This workflow seamlessly integrates simulation, robot learning, and edge computing. It accelerates the process of building, training, and deploying intelligent machines. NVIDIA has launched new Isaac GR00T open models. These models empower robots to comprehend natural language instructions and execute complex, multistep tasks using vision language action reasoning.
In addition, NVIDIA has rolled out Cosmos world models for generating synthetic data and training robots on a large scale. These models assist systems in learning more efficiently and generalizing across different environments. The general availability of the open-source physics engine, Newton 1.0, offers a swift and dependable foundation for dexterous robot manipulation. It features accurate collision detection and realistic object contact.
A notable demonstration features NVIDIA Omniverse developer Umang Chudasama. He integrated NVIDIA NemoClaw with NVIDIA Isaac Sim to guide an autonomous robot using simple natural language commands. There was no manual coding required. NemoClaw translates text instructions like “move two meters forward” into executable Python scripts in real time.
NVIDIA also launched RoboLab, a high-fidelity simulation benchmark built on NVIDIA Isaac and Omniverse simulation technologies. It utilizes photorealistic environments and physics-based modeling to train and test robotic policies on a large scale. This allows researchers to gauge how well behaviors learned in simulation transfer to the real world as tasks become more complex.
Developers no longer need to program robots line by line. They can now simply converse with them. This marks a significant shift towards truly collaborative, language-driven robotics.
Source: https://blogs.nvidia.com/blog/national-robotics-week-2026/