Newton's Classroom - The Physics Class Every Robot Should Attend

Newton Physics Engine represents a breakthrough in robotics simulation, offering an open-source, GPU-accelerated platform that bridges the critical gap between simulated and real-world robot performance. Developed collaboratively by NVIDIA, Google DeepMind, and Disney Research, and managed by the Linux Foundation, Newton is built on NVIDIA Warp and OpenUSD frameworks to provide high-fidelity physics simulation with unprecedented speed and accuracy.

In this livestream, we'll explore Newton's modular architecture supporting multiple physics domains—rigid body dynamics, soft body simulation, cloth modeling, and granular materials. You'll discover how Newton's differentiable physics enables gradient-based optimization for efficient robot learning, and how it can be used with robot learning frameworks like NVIDIA Isaac Lab and MuJoCo playground.

About our livestreams:
Welcome to OpenUSD Insider—a livestream series that brings together developers, researchers, and industry leaders to explore the latest tools, workflows, and innovations shaping the future of OpenUSD. From digital twins and simulation to robotics and industrial applications, each episode features real-world use cases, technical deep dives, and insights from those working on the cutting edge.

Whether you're building complex systems, training autonomous robots, or streamlining simulation pipelines, this series is your front-row seat to the evolving OpenUSD ecosystem.