NVIDIA has announced technologies to supercharge humanoid robot development, including NVIDIA Isaac GR00T N1, the world’s first open, fully-customisable foundation model for generalised humanoid reasoning and skills. Also new are the NVIDIA Isaac GR00T Blueprint for generating synthetic data, as well as Newton, an open-source physics engine purpose-built for developing robots.
GR00T N1 is the first of a family of fully-customisable models that NVIDIA will pretrain and release to worldwide robotics developers — accelerating the transformation of industries challenged by global labour shortages estimated at more than 50 million people.
“The age of generalist robotics is here,” said Jensen Huang, founder and CEO of NVIDIA.
“With NVIDIA Isaac GR00T N1 and new data-generation and robot-learning frameworks, robotics developers everywhere will open the next frontier in the age of AI.”
The GR00T N1 foundation model features a dual-system architecture, inspired by principles of human cognition. System 1 is a fast-thinking action model, mirroring human reflexes or intuition. System 2, on the other hand, is a slow-thinking model for more deliberate and methodical decision-making.
Powered by a vision language model, System 2 reasons about its environment and the instructions it has received to plan actions. System 1 then translates these plans into precise, continuous robot movements. System 1 is trained on human demonstration data and a massive amount of synthetic data generated by the NVIDIA Omniverse platform.
NVIDIA said GR00T N1 can generalise across common tasks — such as grasping, moving objects with one or both arms, and transferring items from one arm to another — or perform multistep tasks that require long context and combinations of general skills. These capabilities can be applied across use cases such as material handling, packaging and inspection.
Developers and researchers can further post-train GR00T N1 with real or synthetic data for a specific humanoid robot or task.
 |
| Source: NVIDIA. Humanoid robots built on GR00T N1. |
In his GTC keynote, Huang demonstrated 1X’s humanoid robot autonomously performing domestic tidying tasks using a post-trained policy built on GR00T N1. The robot’s autonomous capabilities are the result of an AI training collaboration between 1X and NVIDIA.
“The future of humanoids is about adaptability and learning,” said Bernt Børnich, CEO of 1X Technologies.
“While we develop our own models, NVIDIA’s GR00T N1 provides a significant boost to robot reasoning and skills. With minimal post-training data, we fully deployed on NEO Gamma — advancing our mission of creating robots that are not just tools, but companions capable of assisting humans in meaningful, immeasurable ways.”
NEO Gamma is 1X's humanoid robot, envisioned to be a household assistant. Other humanoid developers worldwide with early access to GR00T N1 include Agility Robotics, Boston Dynamics, Mentee Robotics and NEURA Robotics.
NVIDIA also announced a collaboration with Google DeepMind and Disney Research to develop Newton, an open-source physics engine that lets robots learn how to handle complex tasks with greater precision. Built on the NVIDIA Warp framework, Newton will be optimised for robot learning and compatible with simulation frameworks such as Google DeepMind’s MuJoCo and NVIDIA Isaac Lab.
The three companies plan to enable Newton to use Disney’s physics engine.
Google DeepMind and NVIDIA are collaborating to develop MuJoCo-Warp, which is expected to accelerate robotics machine learning workloads by more than 70x and will be available to developers through Google DeepMind’s MJX open-source library, as well as through Newton.
Disney Research will be one of the first to use Newton to advance its robotic character platform that powers next-generation entertainment robots, such as the Star Wars-inspired BDX droids that joined Huang on stage during his GTC keynote.
“The BDX droids are just the beginning. We’re committed to bringing more characters to life in ways the world hasn't seen before, and this collaboration with Disney Research, NVIDIA and Google DeepMind is a key part of that vision,” said Kyle Laughlin, Senior VP at Walt Disney Imagineering Research & Development.
“This collaboration will allow us to create a new generation of robotic characters that are more expressive and engaging than ever before — and connect with our guests in ways that only Disney can.”
NVIDIA and Disney Research, along with Intrinsic, shared an additional collaboration to build OpenUSD pipelines and best practices for robotics data workflows.
NVIDIA further tackled the challenge of providing enough data for training robots. Large, diverse, high-quality datasets are critical for robot development but costly to capture, the company explained, whereas real-world human demonstration data is limited by a person’s day.
The new NVIDIA Isaac GR00T Blueprint for synthetic manipulation motion generation helps address this challenge. Built on Omniverse and NVIDIA Cosmos Transfer world foundation models, the blueprint lets developers generate exponentially large amounts of synthetic motion data for manipulation tasks from a small number of human demonstrations.
Using the first components available for the blueprint, NVIDIA generated 780,000 synthetic trajectories — the equivalent of 6,500 hours, or nine continuous months, of human demonstration data — in just 11 hours. Then, combining the synthetic data with real data, NVIDIA improved GR00T N1’s performance by 40%, compared with using only real data.
To further equip the developer community with training data, NVIDIA is releasing the GR00T N1 dataset as part of a larger open-source physical AI dataset — also announced at GTC and available on Hugging Face.
NVIDIA separately rounded out its automotive portfolio with NVIDIA Halos, a safety system bringing together NVIDIA’s lineup of automotive hardware and software safety solutions with its AI research in autonomous vehicle (AV) safety. Halos spans chips and software to tools and services to help ensure safe development of autonomous vehicles (AVs) from the cloud to the car.
Meanwhile, the new Open Power AI Consortium, which includes energy companies, technology companies and researchers developing related AI applications, took aim at AI adoption in the power sector with open models using curated, industry-specific data. As part of the consortium, EPRI, NVIDIA and Articul8, a member of the NVIDIA Inception programme for startups, are developing a set of domain-specific, multimodal large language models trained on massive libraries of proprietary energy and electrical engineering data from EPRI that can help utilities streamline operations, boost energy efficiency and improve grid resiliency.
On the engineering front, NVIDIA's CUDA-X is bringing accelerated computing to a broad new set of engineering disciplines, including astronomy, particle physics, quantum physics, automotive, aerospace and semiconductor design.
Details
NVIDIA GR00T N1 training data and task evaluation scenarios are now available for download from Hugging Face and GitHub. The NVIDIA Isaac GR00T Blueprint for synthetic manipulation motion generation is also now available as an interactive demo on build.nvidia.com or to download from GitHub.
The NVIDIA DGX Spark personal AI supercomputer, also announced the same day at GTC, provides developers a turnkey system to expand GR00T N1’s capabilities for new robots, tasks and environments without extensive custom programming.
The Newton physics engine is expected to be available later this year.
Hashtag: #GTC2025
No comments:
Post a Comment