News Summary
Leading companies in robot-brain development, industrial and surgical robots, and humanoid innovation are using NVIDIA technology to develop and deploy physical AI at scale.
NVIDIA has introduced Cosmos, world models AI systems that create and simulate digital environments for robots to learn in and ISAAC simulation frameworks, software tools that allow users to design, test, and train robots virtually. The new ISAT GR00TN models offer improved training for intelligent robot behavior.
Collaborative alliances are transforming platform integrations into concrete industry results, including precise electronics assembly, autonomous construction, and AI-powered automation for manufacturers of all sizes.
At GTC, NVIDIA announced new partnerships with the Global Robotics Community and introduced new ISAAC Simulation Frameworks, as well as Cosmos and ISAAC GR00T open models to support next-generation intelligent robot development, training, and deployment.
Industry leaders are leveraging the N-Media platform to advance robotics innovation.
Physical AI has arrived. Every industrial company will become a robotics company, said Jensen Huang, founder and CEO of NVIDIA. NVIDIA’s full-stack platform spanning computing, open models, and software drive frameworks is the foundation for the robotics industry, uniting a worldwide ecosystem to build the intelligent machines that will power the next generation of factories, logistics, transportation, and infrastructure. Yeah, sure.
Testing the World’s Largest Robotic Fleets
As industrial robotics relies more on AI, manufacturers need accurate, high-quality simulations to design, test, and refine systems before deployment.
FANUC, ABB Robotics, Yaskawa, and KUKA, with over 2 million robots installed globally, use NVIDIA Omniverse libraries and ISAAC simulation frameworks within their virtual environments. commissioning tools. These tools allow companies to develop and test complex robotic applications and production lines using accurate digital replicas known as digital twins. The firms are also incorporating NVIDIA Jetson modules, which are AI processors, into their robot controllers to deliver real-time AI on the production line.
Creating Robot Brains for Any Type of Robot
Robotics is evolving from single-task machines to adaptable systems that maintain the accuracy and dependability required by industry. Accomplishing this requires robots to process information more like humans, enabling them to perceive, decide, and act independently. Developers like Field AI and Skild AI are creating general-purpose robot brains with NVIDIA, Cosmos World models for data, and ISAAC simulation frameworks to test their systems. This lets any robot learn new tasks with little extra training. WorldLabs uses ISAAC SIM to test its generative world models, and Generalist AI uses Cosmos to generate synthetic data.
NVIDIA has announced Cosmos 3, the first world foundation model to combine synthetic world creation, vision reasoning, and action simulation. This advancement will accelerate the development of general robot intelligence for multi-faceted environments, paving the way for the next generation of humanoid robots.
Building humanoid robots remains a major challenge in robotics. Replicating human mobility, dexterity, and reasoning requires integrating advanced AI perception and real-time control into safe, reliable, and self-governing systems that can withstand adversaries, such as 1x. Agibot Agility, Agile Robots, Boston Dynamics Figure, Hexagon Robotics, Humanoid, Menti, and Neura Robotics are building the next generation of humanoids using Cosmos World Models. ISAAC SIM and ISASC Lab to accelerate the development and validation of their robots. NVIDIA has announced ISAC Lab 3.0 in early access, enabling faster, larger-scale robot learning on NVIDIA DGX-class infrastructure. Built on the new Newton Physics Engine 1.0 and the NVIDIA PhysX Software Development Kit, it adds multi-physics simulation and enhanced support for complex dexterous manipulation.
Agibot, Humanoid, LG Electronics, Neura Robotics, and Noble Machines use NVIDIA ISAC GR00TN models to speed up industrial deployment of their humanoids. NVIDIA announced that GR00T-N1.7 is available in early access with commercial licensing, offering generalized robotic skills, including advanced dexterous control and production-ready deployments.
During his GTC keynote, Huang reviewed GR00TN2, a next-generation robot foundation model based on Dream Zero research built on a new world action model. GR00TN2 enables robots to complete new tasks in unfamiliar environments more than twice as often as leading vision-language-action models scheduled for release by year-end. GR00TN2 currently ranks first on Malmo Spaces and Robo Arena for generalist robot policies.
These systems are powered by the NVIDIA Jetson Thor robotic computing platform, enabling developers to transition from simulation training to field deployment faster with greater speed, intelligence, and reliability.
Expanding physical AI to healthcare robotics
Healthcare yields a significant opportunity for physical AI. However, using autonomous systems in surgery, imaging, and hospitals requires infrastructure that meets the strictest safety and regulatory compliance standards.
CMR Surgical is using the Cosmos H simulation platform to train and validate the robotic intelligence for its Versius surgical system prior to clinical deployment. Johnson & Johnson MedTech is using Isaac SIM and Cosmos-based post-training workflows to train and validate systems for the Monarch urology platform. Medtronic is exploring NVIDIA, IGX, and Thor to deliver machine-critical exactness and functional safety in surgical robotic systems.
A Global Catalyst for Robotic Innovation
NVIDIA is building an open, unified platform that enables users to design, train, test, and deploy physical AI solutions. This model fosters collaboration within the robotics sector and streamlines scaling real-world integrations.
Supporting The Next Generation Of Physical AI Innovators
NVIDIA wants to make physical AI tools available to everyone, whether you are part of a new startup or the worldwide open source community.
The NVIDIA Inception Program is a global startup incubator with over 40,000 members. It gives robotics innovators access to NVIDIA’s open physical AI stack. Members like Bedrock Robotics, Dexterity AI, Flexicon, Lightwheel, Rivr, Standardbots, Vention, and World Labs receive technical support, powerful computing resources, and opportunities to interact with key robotics partners and customers.
NVIDIA is also working with Hugging Face to integrate ISAAC and GR00T into the LeRobot open-source framework. This partnership brings together NVIDIA’s 2 million robotics developers and Hugging Face’s 13 million AI builders worldwide, helping accelerate open-source robotics development.
Source: NVIDIA and Global Robotics Leaders Take Physical AI to the Real World
