Interested in making your bipedal robots to be athletes? We summarized our RL work to create robust & adaptive controllers for general bipedal skills. 400m-dash, running over terrains/against perturbations, targeted jumping, compliant walking, not a problem for bipeds now.🧵👇

Just arrived in Seattle for #CVPR2024!😎 Catch me at the #HuMoGen workshop this Tuesday, I will be presenting my latest work "Multi-Track Timeline Control for Text-Driven 3D Human Motion Generation"🚀 mathis.petrovich.fr/stmc

We are excited to present our #SIGGRAPH2024 paper, "Interactive Character Control with Auto-Regressive Motion Diffusion Model (AMDM)," on Friday, August 1st, at 2:10 PM in the Mile High 4 room. Project Page: yi-shi94.github.io/amdm_page

Excited to share our latest work! 🤩 Masked Mimic 🥷: Unified Physics-Based Character Control Through Masked Motion Inpainting Project page: research.nvidia.com/labs/par/maske… with: Yunrong (Kelly) Guo, Ofir Nabati, Gal Chechik and Jason Peng. SIGGRAPH Asia ➡️ Hong Kong (ACM TOG). 1/ Read

Excited to share our code for physics-based animation! 🥳 "ProtoMotions -- primitive or fundamental types of movement that serve as a basis for more complex motions.” github.com/NVlabs/ProtoMo… This contains our recent work on MaskedMimic, but some additional pieces. 🧵 1/

Introducing HiLMa-Res: a hierarchical RL framework for quadrupeds to tackle loco-manipulation tasks with sustained mobility! Designed for general learning tasks (vision-based, state-based, real-world data, etc), the robot now can step over stones🐾/navigate boxes📦/dribble⚽.

We’ve seen humanoid robots walk around for a while, but when will they actually help with useful tasks in daily life? The challenge here is the diversity and complexity of real-world scenes. Our new work tackles this problem via 3D visuomotor policy learning. Using data from

Smooth behaviors is vital for successful sim2real transfer of RL policies. This is often achieved with smoothness rewards or low-pass filters, which are not easily differentiable and tend to require tedious tuning. We introduce Lipschitz-Constrained Policies (LCP), a simple and

Our team at NVIDIA is recruiting PhD interns for 2025 to work on a wide range of topics in computer animation! If you want to join us to create some of the most capable and realistic simulated characters, then apply here: nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAEx…

When we bought this humanoid robot, added our control system magic, not just a robot, but a performer. ... 💃🤖 "Bringing rhythm to real robots!" "#BridgeDP_Robotics #AI #Humanoid #EmbodiedAI #Dance #Motion #Control #sim2real

🤖Introducing TWIST: Teleoperated Whole-Body Imitation System. We develop a humanoid teleoperation system to enable coordinated, versatile, whole-body movements, using a single neural network. This is our first step toward general-purpose robots. 🌐humanoid-teleop.github.io

Interested in simulated characters traversing complex terrains? PARC: Physics-based Augmentation with Reinforcement Learning for Character Controllers Project page: michaelx.io/parc/index.html with: Yi Shi, KangKang Yin, and Jason Peng ACM SIGGRAPH 2025 Conference Paper 1/

🚀Introducing GMT — a general motion tracking framework that enables high-fidelity motion tracking on humanoid robots by training a single policy from large, unstructured human motion datasets. 🤖A step toward general humanoid controllers. Project Website:

🚀 Introducing LeVERB, the first 𝗹𝗮𝘁𝗲𝗻𝘁 𝘄𝗵𝗼𝗹𝗲-𝗯𝗼𝗱𝘆 𝗵𝘂𝗺𝗮𝗻𝗼𝗶𝗱 𝗩𝗟𝗔 (upper- & lower-body), trained on sim data and zero-shot deployed. Addressing interactive tasks: navigation, sitting, locomotion with verbal instruction. 🧵 ember-lab-berkeley.github.io/LeVERB-Website/

As we scale up the amount of data used to train humanoid controllers, cleaning motion data is becoming a significant bottleneck. With StableMotion, we can train motion cleanup models directly on raw corrupted motion data, and use it to automatically clean up the whole dataset.

Excited to share our latest work MaskedManipulator (proc. SIGGRAPH Asia ➡️ Hong Kong 2025)! With: Yifeng Jiang, Erwin Coumans 🇺🇦, Zhengyi “Zen” Luo, Gal Chechik, and Jason Peng

I have always been surprised by how few positive samples adversarial imitation learning needs to be effective. With ADD we take this to the extreme! A differential discriminator trained with a SINGLE positive sample can still be effective for a wide range of tasks.