Hi everyone! I’m Jerry Cheng, PhDing @nyu. I work on robotics, simulation, and RL for locomotion + manipulation.
Today I finally shipped my own website:
https://t.co/j3TDsCCIME 🚀
If you do Robotics Research and is also passionate in building a Robotics Startup-Let’s connect!
🧐 Simulation has long promised robot pretraining, but breaks at the moment of real-world deployment.
🚀 Today, we introduce SIM1: the first real-to-sim-to-real paradigm where the generative world becomes the same one as reality.
SIM1 produces simulation data whose execution is directly valid in the physical world, enabling policies trained entirely in simulation to transfer zero-shot, at scale.
📈 This unlocks a new scaling law for robotics: we scale intelligence without scaling real-world data.
✨ Few demonstrations in, real-world policies out.
Simulation is no longer a proxy; it is supervision itself.
https://t.co/Kp1YBe5Gmf
https://t.co/GG2SBQfPpG
About six months ago, I left Stanford to build a robotics company for the home.
At the time, humanoid hype felt like it was getting ahead of its promises but no one had really shown an alternative.
So we’ve been building one. Excited to share more soon.
About six months ago, I left Stanford to build a robotics company for the home.
At the time, humanoid hype felt like it was getting ahead of its promises but no one had really shown an alternative.
So we’ve been building one. Excited to share more soon.
Nvidia's new Kimodo model is seriously impressive - power of foundation models using the current research ideas.
There are some failure cases when I asked it to backflip or stand on one foot but overall insane results.
@HaoL1R Interesting how the low level dynamics is solved! Great work.
Regarding the data collection I am not fully convinced-IL can only be as good as the expert - and if there is already a classical pipeline for underwater grasping, doesn’t this defeat the purpose of having a ml policy
Really excited to release mjviser, a web-based MuJoCo viewer, powered by Viser. It has almost all the features of the native MuJoCo viewer, but runs in your browser. Load and simulate any MuJoCo model with a single uv command 👇
uvx mjviser <model.xml>
@pathak2206 Super cool! But does it also generalize to other tool use?
I feel this demo could also be achieved by non-learning methods using the classical robotics stack. Gripper and rigid attachment may not be good in the long run
It has been quite a journey working with @binghao_huang and colleagues pushing toward scalable tactile sensing. We are now fully open-sourcing a detailed reproduction guide for fabricating the most up-to-date version of our sensor.
From early prototypes to a system that is now reliable, reproducible, and easy to deploy, this effort reflects a lot of persistence across hardware, learning, and full robotic system integration.
FlexiTac is designed to be scalable, flexible, and customizable.
Importantly, it also comes with a simulation stack that shows strong sim-to-real alignment, which we see as critical for making tactile sensing practical in real robotic systems.
We are excited to already see strong interest from colleagues across both academia and industry, with ongoing efforts toward commercialization with our industry partners.
- Project page: https://t.co/zF3yjhAteD
- Walkthrough video: https://t.co/enBmwcWh3x
- Reproduction guide: https://t.co/5OPtMfxUD1
Looking forward to seeing what the community builds with it!
a low-cost 6-DOF robotic arm built to escape simulation
custom gearboxes. modular joints. spherical wrist.
real world robotics isn't learned in gazebo.
it's learned when your PID loop oscillates and your gearbox binds
open source hardware that actually teaches you why things fail
We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.
#NVIDIA just released a whole ecosystem for human(oid) motion and robot learning from human data. 🚀🦾
Data, as we all know, is the key to scaling AI models. To accelerate the field of Embodied AI, we have open-sourced a full stack of models and tools to capture, generate, retarget, and simulate human(oid) motion data at scale, along with a massive high-quality dataset and a standard human skeletal representation, SOMA, to make them all seamlessly communicate with each other.
The entire suite is available under the Apache 2.0 license.
1️⃣ SOMA: A universal interface to unify all parametric human body models (SOMA-shape, SMPL, MHR, etc.) into a standard skeletal representation, eliminating the need for custom adapters or model-specific retargeting.
🔗 https://t.co/Xrg672T7Nu
2️⃣ Kimodo: High-fidelity, controllable text-to-motion generation for both humans and humanoid robots.
🔗 https://t.co/2cQKAPfvEU
3️⃣ GEM: A global human pose estimation method from in-the-wild videos, natively compatible with SOMA.
🔗 https://t.co/pV0043jwcO
4️⃣ Bones-SEED: A massive dataset of 150k+ motions in SOMA format, including data already retargeted for the Unitree G1, created with our partners at Bones Studio.
🔗 https://t.co/wxfyZ7S9TJ
🔗 https://t.co/oM5rIMdRi8
5️⃣ SOMA Retargeter: A dedicated tool for seamless motion retargeting from the SOMA skeleton to the Unitree G1.
🔗 https://t.co/jg4DUjWcnw
6️⃣ ProtoMotions: Our high-performance simulation framework for training digital human(oid)s via RL, now with native SOMA support.
🔗 https://t.co/K1zsGEdl5S
This is just the beginning, and we have much more in the pipeline. Excited to see what the community builds next!
#NVIDIA #GTC #GTC2026 #Robotics #EmbodiedAI #PhysicalAI @NVIDIAAI