{"id":3785,"date":"2024-11-06T16:42:42","date_gmt":"2024-11-06T16:42:42","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2024\/11\/06\/nvidia-advances-robot-learning-and-humanoid-development-with-new-ai-and-simulation-tools\/"},"modified":"2024-11-06T16:42:42","modified_gmt":"2024-11-06T16:42:42","slug":"nvidia-advances-robot-learning-and-humanoid-development-with-new-ai-and-simulation-tools","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2024\/11\/06\/nvidia-advances-robot-learning-and-humanoid-development-with-new-ai-and-simulation-tools\/","title":{"rendered":"NVIDIA Advances Robot Learning and Humanoid Development With New AI and Simulation Tools"},"content":{"rendered":"
\n\t\t<\/span> <\/span> <\/span><\/span><\/p>\n

Robotics developers can greatly accelerate their work on AI-enabled robots, including humanoids<\/a>, using new AI and simulation tools and workflows that NVIDIA revealed this week at the Conference for Robot Learning (CoRL<\/a>) in Munich, Germany.<\/p>\n

The lineup includes the general availability of the NVIDIA Isaac Lab<\/a> robot learning framework; six new humanoid robot learning workflows for Project GR00T<\/a>, an initiative to accelerate humanoid robot development; and new world-model development tools for video data curation and processing, including the NVIDIA Cosmos tokenizer<\/a> and NVIDIA NeMo Curator<\/a> for video processing.<\/p>\n

The open-source Cosmos tokenizer provides robotics developers superior visual tokenization by breaking down images and videos into high-quality tokens with exceptionally high compression rates. It runs up to 12x faster than current tokenizers, while NeMo Curator provides video processing curation up to 7x faster than unoptimized pipelines.<\/p>\n

Also timed with CoRL, NVIDIA presented 23 papers and nine workshops related to robot learning and released training and workflow guides for developers. Further, Hugging Face and NVIDIA announced<\/a> they\u2019re collaborating to accelerate open-source robotics research with LeRobot, NVIDIA Isaac Lab and NVIDIA Jetson<\/a> for the developer community.<\/p>\n

Accelerating Robot Development With Isaac Lab\u00a0<\/b><\/h2>\n

NVIDIA Isaac Lab is an open-source<\/a>, robot learning<\/a> framework built on NVIDIA Omniverse<\/a>, a platform for developing OpenUSD<\/a> applications for industrial digitalization and physical AI<\/a> simulation.<\/p>\n

Developers can use Isaac Lab to train robot policies at scale. This open-source unified robot learning framework applies to any embodiment \u2014 from humanoids to quadrupeds to collaborative robots \u2014 to handle increasingly complex movements and interactions.<\/p>\n

Leading commercial robot makers, robotics application developers and robotics research entities around the world are adopting Isaac Lab, including 1X, Agility Robotics<\/a>, The AI Institute, Berkeley Humanoid<\/a>, Boston Dynamics<\/a>, Field AI<\/a>, Fourier<\/a>, Galbot<\/a>, Mentee Robotics<\/a>, Skild AI, Swiss-Mile, Unitree Robotics and XPENG Robotics.<\/p>\n

Project GR00T: Foundations for General-Purpose Humanoid Robots\u00a0<\/b><\/h2>\n

Building advanced humanoids is extremely difficult, demanding multilayer technological and interdisciplinary approaches to make the robots perceive, move and learn skills effectively for human-robot and robot-environment interactions.<\/p>\n

Project GR00T<\/a> is an initiative to develop accelerated libraries, foundation models and data pipelines to accelerate the global humanoid robot developer ecosystem.<\/p>\n

\"\"<\/p>\n

Six new Project GR00T workflows<\/a> provide humanoid developers with blueprints to realize the most challenging humanoid robot capabilities. They include:<\/i><\/p>\n