CS PhD at Stanford
Stanford, CA
Joined January 2021
- Tweets189
- Following254
- Followers2.5K
- Likes277
Pinned Tweet
Excited to release Minimalist Compliance Control! We achieve robust, compliant robot interaction across robot arms, dexterous hands, and humanoids, with NO force sensors or learning. If you’re wondering what remains, please see the thread below😉
Website: minimalist-compliance-contro…
The speed, subconscious movements, and cross-embodiment results are super impressive. Big congrats to the team 🔥🔥🔥
Haochen Shi retweeted
💡💡💡Wednesday’s Thing is coming... with a humanoid mode.
🤖 Meet Handroid: a 33 cm, 2.05 kg, 27-DoF desktop robot that reconfigures the same hardware into a five-finger dexterous hand or a humanoid.
🖐️ As a 20-DoF hand, Handroid picks up paper, pours water, grasps a ball, stacks cups, removes a glove, and reorients a cube in-hand. Its stack combines Vision Pro teleoperation, Diffusion Policy grasping, an1d RL-based in-hand reorientation.
🚶 Reconfigure it: five finger modules become a head, two arms, and two legs. Handroid walks, turns, sidesteps, squats, pushes boxes, and even does push-ups and pull-ups. We validate RL tracking and velocity control in simulation, and deploy keyframe motions on hardware.
🖐️🚶🖐️ In one long-horizon demo, Handroid starts docked to a Franka arm as a hand, becomes a humanoid, undocks, moves around an obstacle, and pushes a box within reach. It then returns to hand form, re-docks, picks up a bottle, and places it in the box.
💡 A robot does not have to be defined from the outset as only a hand, a pair of legs, or a body locked into one fixed form. Its joints can be reassigned, its modules recombined, and the same hardware can take on entirely different roles as tasks change. When morphology becomes a reconfigurable, reusable resource rather than a fixed boundary on capability, new possibilities emerge. Handroid is not the final answer, but a starting point. We hope it inspires researchers to rethink what robots can be and what they can do.
🌐 handroid.org
📄 arxiv.org/abs/2607.16187
#Robotics
This is exactly the kind of "transformer" I imagine. Creatively extending the otherwise limited workspace of a robot arm in a way that just makes sense. A perfect example of a brilliant idea + amazing execution. Enjoyed collaborating with the team a lot!
🤖 A robot that transforms between humanoid and dexterous hand?
Introducing Handroid, a reconfigurable robot with a shared 27-DoF body 🖐️→🚶→🖐️:
• The same joints, different roles
• Expanded robot task space
• Shared sensing and control interfaces
handroid.org
Haochen Shi retweeted
A person walks around campus for 5 hours with cameras.
That's it. That's the training data.
The result? A humanoid robot that traverses unseen buildings, crowds, and glass walls — zero robot data, zero finetuning.
EgoNav is here.
egonav.weizhuowang.com/
None of these behaviors were pre-programmed:
• Waiting for a door to open before entering
• Steering around glass walls invisible to depth sensors
• Yielding to pedestrians and resuming
• Re-routing when furniture is rearranged
All emerged from 5 hours of a human walking around. The prior is real. (1/6)
#Humanoid #Robotics #DiffusionModel #EgoNav
Haochen Shi retweeted
🤲Tactile sensing is powerful for robot manipulation, but hardware is still difficult to access, reproduce, and scale.
🎯That’s why we built FlexiTac: an open-source, low-cost, and scalable tactile sensing solution designed for real robotic systems.
• Project page: flexitac.github.io/
We hope FlexiTac can help democratize tactile sensing for robotics research. (1/n)
Haochen Shi retweeted
I’m so tired of writing rebuttals to this kind of “lack of novelty” review: “This paper trivially combines A, B, and C, so the algorithmic novelty is limited.”
Technically, most (if not all) robotics papers are convex combinations of existing ideas.
I still deeply appreciate A+B+C papers—especially when they deliver:
- New capabilities: the “trivial combination” unlocks behaviors we simply couldn’t achieve before
- Sensible & organic design: A+B+C is clearly the right composition—not some arbitrary A′+B+C′
- Nontrivial interactions: careful analysis of the dynamics, coupling, or failure modes between A, B, C
- Rehabilitating old ideas: A was dismissed for years, but paired with modern B/C, it suddenly works—and teaches us why
- System-level & "interface" insight: the contribution is not any single piece, but how the pieces talk to each other
- Scaling laws or regimes: identifying when/why A+B+C works (and when it doesn’t)
- Engineering clarity: making something actually work robustly in the real world is not “trivial”
- New problem formulations: sometimes the real novelty is in the reformulation—only under this view does A+B+C make sense.
Maybe worth keeping these in mind when reviewing the next A+B+C paper : )
Haochen Shi retweeted
Can we learn whole-body mobile manipulation directly from human demonstrations?
Introducing Whole-Body Mobile Manipulation Interface (HoMMI)
Egocentric + UMI, 0 teleop -> bimanual & whole-body manipulation, long-horizon navigation, active perception
hommi-robot.github.io
Haochen Shi retweeted
We find that RL post-training can substantially improve BC policies without teaching them anything fundamentally new.
So what is RL doing? In DICE-RL, it contracts a broad behavior prior toward high-value modes. (1/n)
zhanyisun.github.io/dice.rl.…
Haochen Shi retweeted
1/ World models are getting popular in robotics 🤖✨
But there’s a big problem: most are slow and break physical consistency over long horizons.
2/ Today we’re releasing Interactive World Simulator:
An action-conditioned world model that supports stable long-horizon interaction.
3/ Key result:
✅ 10+ minutes of interactive prediction
✅ 15 FPS
✅ on a single RTX 4090🔥
4/ Why this matters: it unlocks two critical robotics applications:
🚀 Scalable data generation for policy training
🧪 Faithful policy evaluation
5/ You can play with our world model NOW at yixuanwang.me/interactive_wo…. NO git clone, NO pip install, NO python. Just click and play!
NOTE ⚠️
ALL videos here are generated purely by our model in pixel space! They are **NOT** from a real camera
More details coming 👇 (1/9)
#Robotics #AI #MachineLearning #WorldModels #RobotLearning #ImitationLearning
As long as your robot uses modern servos (e.g., Dynamixel) or QDD motors, even without current sensing or current control, our method still works! It is truly embodiment-agnostic and democratizes compliant control for low-cost robots. Try it out here: github.com/real-stanford/min…
Excited to release Minimalist Compliance Control! We achieve robust, compliant robot interaction across robot arms, dexterous hands, and humanoids, with NO force sensors or learning. If you’re wondering what remains, please see the thread below😉
Website: minimalist-compliance-contro…
To evaluate our compliance controller against learning-based methods, we compare against two RL-based baselines, UniFP and FACET.
Everything is open-sourced!
Website: minimalist-compliance-contro…
Paper: arxiv.org/abs/2603.00913
Code: github.com/real-stanford/min…
This work is done with @Songbo_Hu_ @YifanHou2 @KenWangWeizhuo, and my amazing advisors @ckarenliu @SongShuran