{"id":4291,"date":"2025-09-30T13:41:47","date_gmt":"2025-09-30T13:41:47","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2025\/09\/30\/into-the-omniverse-open-source-physics-engine-and-openusd-advance-robot-learning\/"},"modified":"2025-09-30T13:41:47","modified_gmt":"2025-09-30T13:41:47","slug":"into-the-omniverse-open-source-physics-engine-and-openusd-advance-robot-learning","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2025\/09\/30\/into-the-omniverse-open-source-physics-engine-and-openusd-advance-robot-learning\/","title":{"rendered":"Into the Omniverse: Open-Source Physics Engine and OpenUSD Advance Robot Learning"},"content":{"rendered":"
\n\t\t<\/span> <\/span> <\/span><\/span><\/p>\n

Editor\u2019s note: This blog is a part of <\/i>Into the Omniverse<\/i><\/a>, a series focused on how developers, 3D practitioners and enterprises can transform their workflows using the latest advances in OpenUSD and NVIDIA Omniverse.<\/i><\/p>\n

Building robots that can effectively operate alongside human workers in factories, hospitals and public spaces presents an enormous technical challenge. These robots require humanlike dexterity, perception, cognition and whole-body coordination to navigate unpredictable real-world environments in real time.<\/p>\n

A \u201csim-first\u201d approach<\/a> unlocks these critical skills by enabling parallel training of hundreds or thousands of robot instances using real robot-captured data and synthetic data<\/a> in simulation environments. Universal Scene Description<\/a> (OpenUSD) provides the foundational framework for this advanced robot development, serving as a scalable, interoperable data standard that enables developers to build physically accurate virtual worlds where robots can practice and perfect their skills before transferring them to real-world applications.<\/p>\n

Accelerating Physical AI Development<\/b><\/h2>\n

NVIDIA announced at the Conference on Robot Learning this week groundbreaking advances in open-source physics simulation, open foundation models and development frameworks, including:<\/p>\n

Newton Physics Engine<\/b><\/a>: While robots learn faster and safer in simulation<\/a>, humanoid robots<\/a> \u2014 with complex joints, balance and movements \u2014 are pushing today\u2019s physics engines to the limit.<\/p>\n

Codeveloped by Google DeepMind, Disney Research and NVIDIA, and managed by the Linux Foundation<\/a>, Newton<\/a> is an open-source, GPU-accelerated physics engine to advance robot learning.<\/p>\n

Built on NVIDIA Warp<\/a> and OpenUSD<\/a>, Newton enables robots to learn complex tasks more precisely while working seamlessly with robot learning<\/a> frameworks like MuJoCo Playground and NVIDIA Isaac Lab<\/a>.<\/p>\n

\"\"<\/p>\n

Tune in to an upcoming <\/i>livestream for a Newton beta demonstration<\/i><\/a> to learn about Newton\u2019s core features and how to get started on Newton with NVIDIA Isaac Lab.<\/i>\u00a0\u00a0<\/i><\/p>\n

Isaac GR00T N1.6<\/b><\/a>: To perform humanlike tasks in the physical world, humanoids must understand ambiguous instructions and navigate unforeseen scenarios. The latest Isaac GR00T N1.6 open robot foundation model, available soon on Hugging Face, integrates NVIDIA Cosmos Reason<\/a>, an open reasoning vision language model built for physical AI<\/a>. Cosmos Reason serves as the robot\u2019s deep-thinking brain and transforms vague instructions into step-by-step action plans using prior knowledge, common sense and physics understanding.<\/p>\n

NVIDIA Isaac Lab<\/b><\/a>:<\/b> The latest version of Isaac Lab, an open-source, modular robot learning framework built on NVIDIA Isaac Sim and OpenUSD, is now available as an early developer release. Version 2.3 brings a host of new features to robotics researchers and developers, including advanced whole-body control and expanded teleoperation for data collection.<\/p>\n

OpenUSD\u2019s interoperability ensures these advanced physics simulations, foundation models and learning frameworks work together seamlessly, enabling developers to build unified robot learning<\/a> pipelines that scale across different platforms and deployment scenarios.<\/p>\n

See How Developers Are Accelerating Robot Learning<\/b><\/h2>\n

Leading humanoid and robotics developers, including Agility Robotics, Lightwheel, Mentee and Universal Robots are adopting simulation technologies and libraries to accelerate physical AI development and deployment.<\/p>\n