Sign in

Stone Tao

@stonet2000.bsky.social
3.1K followers 208 following 163 posts

simulation & research @physical_int stoneztao.com

PostsRepliesMedia
Stone Tao @stonet2000.bsky.social · 19/09/2026
Met some absolute legends of the simulation community, an incredible conference hosted by nvidia and google in london!
150
Stone Tao @stonet2000.bsky.social · 07/09/2026
will be in London Sept 14-17th. If you want to chat let me know! Looking to learn about the UK robotics space and see how our research @ Pi can help!
061
Stone Tao @stonet2000.bsky.social · 31/07/2026
Just over a month later, I have now joined Physical Intelligence full time! Wrote a bit about why I'm excited to research simulation at Pi stoneztao.substack.com/p/joining-ph...
stoneztao.substack.com
Joining Physical Intelligence to Work on Simulation
What makes researching simulation at Pi exciting!
4271
Stone Tao @stonet2000.bsky.social · 16/06/2026
just arrived in SF and excited to begin my research internship at physical intelligence working on simulation!
0100
Stone Tao @stonet2000.bsky.social · 10/06/2026
For those unaware I had previously told many privately that I planned to move on after I graduate. But after learning the number of new groups / researchers that planned to use ManiSkill and receiving some great feedback, I figured I should do everything I can to set up ManiSkill for the long term.
020
Stone Tao @stonet2000.bsky.social · 10/06/2026
I reverted my decision to hand off ManiSkill to someone else. A ManiSkill 4 is now in progress with lots of improvements (focusing on robot manipulation) If you planned to not use ManiSkill because it might not be maintained once I graduate, I hope this convinces you otherwise!
140
Stone Tao @stonet2000.bsky.social · 08/06/2026
- clean out the main repo and split off some parts to new repos under the new github organization (eg robot assets/python classes and some baseline algorithms) - better support for policy evals
010
Stone Tao @stonet2000.bsky.social · 08/06/2026
Additionally there’s a new roadmap that is reflective what I’ve been planning to improve in the future. Some plans include - switching to Newton/Mujoco warp backend - a big redo of some apis inherited from ManiSkill2 based on feedback and my own pain points using maniskill
120
Stone Tao @stonet2000.bsky.social · 08/06/2026
The ManiSkill project now has a new home under a new mani-skill github organization, moving out of the old UCSD lab org. Long overdue and done as part of my plans for longer term maintainance as well as more organized community driven development github.com/mani-skill
github.com
ManiSkill
Community for developing the Maniskill robotics simulation framework - ManiSkill
151
Stone Tao @stonet2000.bsky.social · 08/06/2026
- clean out the main repo and split off some parts to new repos under the new github organization (eg robot assets/python classes and some baseline algorithms) - better support for policy evals
010
Stone Tao @stonet2000.bsky.social · 08/06/2026
Additionally there’s a new roadmap that is reflective what I’ve been planning to improve in the future. Some plans include - switching to Newton/Mujoco warp backend - a big redo of some apis inherited from ManiSkill2 based on feedback and my own pain points using maniskill
110
Stone Tao @stonet2000.bsky.social · 02/06/2026
and in the off chance you are a saber fencer, I brought my fencing gear (and will be at the Salle Auriol Seattle fencing club) 🤺
010
Stone Tao @stonet2000.bsky.social · 02/06/2026
in Seattle for 2 weeks finishing up my work at AI2 (at their extremely beautiful office). lmk if you want to meet!
130
Stone Tao @stonet2000.bsky.social · 13/05/2026
First blog post up on Robotics Simulation Infrastructure! I give a high-level overview, followed by an elementary example of better infrastructure for pose management. stoneztao.substack.com/p/robotics-s...
stoneztao.substack.com
Robotics Simulation Infrastructure Part 1
First part of a blog series diving into the world of robotics simulation and machine learning
0375
Stone Tao @stonet2000.bsky.social · 02/04/2026
Wasn't at GTC but a sneak peak of some of my internship work at NVIDIA was! (sim2real cable routing presented by my manager Yashraj) www.youtube.com/watch?v=d5wm...
youtube.com
AI Research Breakthroughs from NVIDIA Research (Hosted by Karoly of Two Minute Papers) | NVIDIA GTC
YouTube video by NVIDIA Developer
010
Stone Tao @stonet2000.bsky.social · 02/04/2026
some points of comparisons could include e.g. manager based configs (mjlab/isaac lab) vs direct style (maniskill/robosuite/also isaac lab), wrappers vs no gym wrappers, features for simplifying parallelization, visualization tooling and more
130
Stone Tao @stonet2000.bsky.social · 02/04/2026
would people be interested in a blog series on "simulation infrastructure" where I look at popular robotics sim tools (e.g. mjlab, isaac...) and explain the pros and cons of their design decisions. The series would cover what I believe to be key pillars of good, usable, sim code
2191
Stone Tao @stonet2000.bsky.social · 01/11/2025
I will be in SF from November 10-17 🌉 If you work on something interesting and want to meet up let me know! Or if there’s a fun event i’d love some invites 😃 I’m also looking to visit robotics companies (especially startups!), if you have time lmk if I can visit!
010
Stone Tao @stonet2000.bsky.social · 19/09/2025
1/1 neurips submission accepted! Will share more details later, covers an often overlooked aspect of on policy RL training when scaling to large scale parallelized environments
020
Reposted by Stone Tao
Nathan Lambert @natolambert.bsky.social · 19/09/2025
Rumors going around are that NeurIPS PCs had to reject ~400 papers because of venue constraints that were in the accept pile. Hard things like this happen but the PCs and conference should explicitly tell all the authors who had this happen to them. Or create something like accept w/o presentation.
1191
Stone Tao @stonet2000.bsky.social · 17/09/2025
what’s your next blog going to be? i’ve been also meaning to pick it up again but writers block lol
110
Stone Tao @stonet2000.bsky.social · 09/09/2025
It works best if the object meshes you use for diff. rendering have a corresponding segmentation mask. If there are occlusions then the default option of using SAM2 to generate segmentations will perform a little worse. That being said even if the mask has some gaps the optimization can work
010
Stone Tao @stonet2000.bsky.social · 09/09/2025
Big thanks to Linghao Chen (author of EasyHEC) for helping out. I simply made some examples and simplified some code for reproducibility/readability.
030
Stone Tao @stonet2000.bsky.social · 09/09/2025
Opensourcing a tool to calibrate camera extrinsics painlessly in a minute, no checkerboards! It's based on EasyHEC, using diff. rendering to optimize extrinsics given object meshes+poses. Crazy that even a piece of paper works too. Code: github.com/StoneT2000/s... (paper example in next post)
2247
Stone Tao @stonet2000.bsky.social · 20/08/2025
Big thanks to Jeremy Morgan (PhD @ USC) for helping massively accelerate the IK solvers in ManiSkill! Open source really helps grow this project (and i get to learn tons of new things like IK optimization) github.com/haosulab/Man...
github.com
[Feature] Faster IK based controllers using levenberg-marqurdt and align pd ee pos controllers in gpu/cpu sims by StoneT2000 · Pull Request #1213 · haosulab/ManiSkill
Cleaned up version of #1148 to close #955. Updates the original code to permit various other pd ee controller options with the new solver. LM solver is now the default. Users can modify the IK sol...
030
Stone Tao @stonet2000.bsky.social · 19/07/2025
ManiSkill3 simple example, Mujoco Playground, and the DextrahRGB paper all do very similar things. I still think transfer techniques are useful, i’d think of these sim2real attempts as just explorations in the large scale visual DR space with parallel rendering. Both could be used together probably
010
Stone Tao @stonet2000.bsky.social · 19/07/2025
see section D of maniskill3 paper: arxiv.org/abs/2410.00425 tldr: Green screen background, segment out robot and cube in sim and only render that. Large scale visual domain randomization (robot, cube textures, camera parameters) very simple trick that is fairly reproducible for simple tasks
arxiv.org
ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
Simulation has enabled unprecedented compute-scalable approaches to robot learning. However, many existing simulation frameworks typically support a narrow range of scenes/tasks and lack features crit...
130
Stone Tao @stonet2000.bsky.social · 19/07/2025
might be the first time I made it personally on the trending developers list on github! One of the contributing factors being the lerobot sim2real code
1170
Reposted by Stone Tao
Chris Paxton @cpaxton.bsky.social · 05/07/2025
Combining sim to real with cheap hardware like the so-100 really is making cutting edge robot learning so accessible and its wonderful
5487
Stone Tao @stonet2000.bsky.social · 05/07/2025
That doesn’t seem to make sense, you don’t need a RGBD camera, phone works fine. Happy to address it more if details can be shared on github if needed
110
Stone Tao @stonet2000.bsky.social · 05/07/2025
what bug did you encounter with OpenCVCameraConfig?
110
Stone Tao @stonet2000.bsky.social · 04/07/2025
For papers that aren’t ready / i think should be rejected due to many weaknesses, i often struggle to put anything in strengths and tend to just say some general thing. I only then elaborate more detailed (i hope) in the weaknesses section
110
Reposted by Stone Tao
Nathan Lambert @natolambert.bsky.social · 04/07/2025
My latest post: The American DeepSeek Project Build fully open models in the US in the next two years to enable a flourishing, global scientific AI ecosystem to balance China's surge in open-source and an alternative to building products ontop of leading closed models. buff.ly/kvJQE3I
1346
Stone Tao @stonet2000.bsky.social · 04/07/2025
Awesome to see people reproducing the accessible LeRobot zero-shot sim2real project! Trained for just 90 min in ManiSkill and deployed directly in real. Sim2real is not easy, but very rewarding when it works Original post by Jianwei Zhang on LinkedIn www.linkedin.com/posts/jianwe...
0296
Stone Tao @stonet2000.bsky.social · 28/06/2025
Excited to announce that I will be interning at @nvidia research this summer on robotics/embodied AI! I’ll be in seattle for the summer, let me know if you want to meet up and chat! 🦾
0100
Stone Tao @stonet2000.bsky.social · 22/06/2025
June 25 EEB 248 @ 3pm - Towards Embodiment Scaling Laws in Robot Locomotion by @BoAi0110 June 25 OHE 122 @ 11am - ImVR: Immersive VR Teleoperation System for General Purpose by Yulin Liu
020
Stone Tao @stonet2000.bsky.social · 22/06/2025
Rest of my lab has various presentations at RSS2025, please check out their awesome work! June 25 EEB 248 @ 10am - Hardware Optimization for In-Hand Rotation by K. Fay
130
Stone Tao @stonet2000.bsky.social · 22/06/2025
~1 hour give or take maybe 20 mins (depends how close your camera is technically), and only 8gb of GPU memory! someone even reproduced our sim2real tutorial with lerobot during the lerobot hackathon with a google colab gpu and a macbook for robot deployment
140
Stone Tao @stonet2000.bsky.social · 22/06/2025
The sim2real demo had some mixed success, hampered primarily by the lighting conditions of the outdoors. At least it worked sometimes! Hindsight says that despite the weather, low-cost nature, only 1 hour of training, anything working is a miracle
192
Stone Tao @stonet2000.bsky.social · 20/06/2025
I’ll be at #RSS2025 from June 21 to June 23! I’ll be giving a presentation on ManiSkill3 on June 21, 5:30 PM We will also have two live demo sessions, on June 21, 12:30-2:00PM and 6:30-8:00PM. Swing by to see live demos of zero shot RGB sim2real, cool sim demos, and VR teleop!
031
Reposted by Stone Tao
Phillip Isola @phillipisola.bsky.social · 15/06/2025
Our computer vision textbook is now available for free online here: visionbook.mit.edu We are working on adding some interactive components like search and (beta) integration with LLMs. Hope this is useful and feel free to submit Github issues to help us improve the text!
visionbook.mit.edu
Foundations of Computer Vision
The print version was published by
311633
Stone Tao @stonet2000.bsky.social · 15/06/2025
Email so far ahead is interesting. Is this just substack emails?
120
Stone Tao @stonet2000.bsky.social · 15/06/2025
the amount of cope on the other site is crazy
030
Reposted by Stone Tao
Thomas Wood @advanced-eschatonics.com · 15/06/2025
it's happening.gif @stonet2000.bsky.social
141
Reposted by Stone Tao
Tzu-Mao Li @tzumaoli.bsky.social · 14/06/2025
Rendering nerds! Check out our latest work "Vector-Valued Monte Carlo Integration Using Ratio Control Variates" that has just gotten the best paper award at SIGGRAPH 2025. This paper presents a method that reduces variance of a wide range of rendering and diff. rendering tasks with negligible cost.
79022
Stone Tao @stonet2000.bsky.social · 14/06/2025
code and tutorial open sourced here! github.com/StoneT2000/l...
github.com
GitHub - StoneT2000/lerobot-sim2real: lerobot sim2real code
lerobot sim2real code. Contribute to StoneT2000/lerobot-sim2real development by creating an account on GitHub.
010
Stone Tao @stonet2000.bsky.social · 13/06/2025
Code/tutorial released for🤗LeRobot zero-shot visual sim2real Reinforcement Learning! Train your SO100 robot in ManiSkill's fast simulator+renderer💨 and deploy a cube picking model zero-shot. Just in time for the LeRobot Hackathon, happy hacking! github.com/StoneT2000/l...
github.com
GitHub - StoneT2000/lerobot-sim2real: lerobot sim2real code
lerobot sim2real code. Contribute to StoneT2000/lerobot-sim2real development by creating an account on GitHub.
090
Stone Tao @stonet2000.bsky.social · 12/06/2025
ah i didn’t know! I bought this one from WowRobo, they provided these clamps
110
Stone Tao @stonet2000.bsky.social · 12/06/2025
the green color haha?
110
Stone Tao @stonet2000.bsky.social · 12/06/2025
longer uncut video 17/18 successes increased the spawn area size and tuned it to be less occluded. Works well!
051