Rui Chen (@RuiChen_Rob) header
Rui Chen (@RuiChen_Rob) profile picture

Ph.D. candidate @CMU_Robotics with @ICL_at_CMU. Constrained Control/Optimization/Learning. @SCSatCMU @UMich

Pittsburgh, PA
Joined February 2024
Photo by Rui Chen (@RuiChen_Rob)Photo by Rui Chen (@RuiChen_Rob)Photo by Rui Chen (@RuiChen_Rob)
Rui Chen retweeted
Excited to share our recent work on EgoX! It turns out that with cross-embodiment pretraining and a little “dreaming,” robots with very different bodies can learn manipulation skills from each other’s demos!
Introducing EgoX: a cross-embodiment learning framework that trains manipulation policies across diverse embodiments using data from humans, physical robots, and simulated robots. We cover highly heterogeneous embodiments: humans, physical and simulated humanoids, wheeled bimanual robots, tabletop arms, and quadrupeds. Their end effectors include three- to five-fingered dexterous hands, two-fingered and parallel-jaw grippers, and a fixed tool. One recipe, any embodiment. egox-dream.github.io/
2
9
391
1/4 Can we scale safety analysis to high-dimensional robotic systems? At #CoRL2026, we presented λ-Reachability, an approach to learning a safety value function that predicts future safety violations for high-dimensional tasks. Watch it flag impending falls and obstacle hits on a real Unitree G1. 🤖 We derive a stochastic multi-step safety Bellman operator that induces a contraction mapping and whose fixed point approaches the undiscounted safety value under mild conditions. 🎉 Project website intelligent-control-lab.gith… 📖 Paper openreview.net/forum?id=sIYf… 📦 Repo github.com/intelligent-contr…
3
7
1
52
2,714
3/4 During training, we sample rollout segments and use the worst safety signals in those segments to update the prediction at the start. λ controls how far ahead we look. Longer segments let each update propagate information about future safety violations back to earlier states.
1
1
91
4/4 Our CoRL paper covers the method and the full results from our G1 balance and collision experiments. 📄 openreview.net/forum?id=sIYf… Joint work by @RuiChen_Rob , @ShangtaoLi , @YifanSun98 , and @ChangliuL at @CMU_Robotics .
1
127
If youre a roboticist worried about how frontier models are going to eat all of robotics, I recommend stopping by the RoboColliseum setup at the BitRobot booth. It made me feel a lot better. Here's GPT6 astra folding a towel
48
33
11
399
47,129
Just started open-sourcing ART-Glove! github.com/linchangyi1/ART-G… Mechanical files and assembly guide are available now. PCB, FPCB, code, and video guide are coming next! I’ll also bring the glove to the CoRL demo session on Monday. Come check it out!
Manipulation happens through surfaces. To understand contact-rich dexterous interaction, motion alone is not enough. We also need to know surface properties and contact state. Excited to share ART-Glove, an articulated tactile glove that captures contact-grounded information while preserving human dexterity. It provides: - Known Geometry: 16 rigid functional surfaces - Surface Motion: 22 anatomically aligned joints - Tactile Contact: 2048 piezoresistive taxels Huge thanks to my advisor Ding @zhao__ding, and to Yuxiang @yxyang1995, Maria @bauzavillalonga, Marissa, and Peide @peide_huang for the valuable advice and discussions. Paper: arxiv.org/abs/2606.16370
10
21
3
177
8,342
Introducing LocoTouch: Quadrupedal robots equipped with tactile sensing can now transport unsecured objects — no mounts, no straps. The tactile policy transfers zero-shot from sim to real. Core Task-Agnostic Features: 1. High-fidelity contact simulation for distributed tactile sensors 2. Adaptive gait reward for symmetric, frequency-adaptive locomotion Website: linchangyi1.github.io/LocoTo… Paper: arxiv.org/abs/2505.23175
1
36
7
127
38,898
🎉 Congrats to Prof Changliu Liu @ChangliuL on IEEE RAS Early Academic Career Award in Robotics and Automation "for significant advancements in intelligent robot control and learning that enables safe and efficient human-robot collaborations"! @CMU_Robotics #ICRA2025
4
11
771
[1/7] Teaching dexterous robot hands to perform functional grasps usually needs hours of teleoperation, manual labeling, or pre-scanning object meshes. Not anymore. 🔥We are excited to introduce Web2Grasp that learns functional multi-finger grasps straight from web images of human hand-object interactions (HOI). No human demos. No object scans. Just web images. 👉web2grasp.github.io/ @CMU_Robotics @CarnegieMellon
1
26
4
76
9,269
🚀 New Research on Human-Robot Interaction! 🤖 How can humanoid robots communicate beyond words? Our framework, EMOTION, leverages Large Language Models (LLMs) to dynamically generate expressive gestures, enhancing non-verbal communication in robots. 🤯 Our experiments show that EMOTION can generate various expressive gestures from only TWO examples and match human-generated gestures in understandability & naturalness! 🔍 What’s inside? ✅ LLM-powered motion generation ✅ Human feedback to refine gestures (EMOTION++) ✅ 10 expressive gestures generated and evaluated (thumbs-up, stop, jazz-hands & more!) 📜 Read the full paper: arxiv.org/abs/2410.23234 🎬 Watch the video: machinelearning.apple.com/re… Let’s bring robots closer to human-like interactions! What gestures would you like to see next? 👇 Huge kudos to the amazing team at Apple that made this work @Yuhan_Hu_, Nataliya Nechyporenko, @talking_kim, @waltertalbott, @jian_zhang_. #Robotics #HRI #LLMs #HumanRobotInteraction #GestureGeneration #SocialRobots
15
81
8
380
26,765
🚀 Can we make a humanoid move like Cristiano Ronaldo, LeBron James and Kobe Byrant? YES! 🤖 Introducing ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills Website: agile.human2humanoid.com/ Code: github.com/LeCAR-Lab/ASAP
44
187
75
1,019
260,073
🚀 Our work is featured in the latest @ARM_Robotics article on the Digital Framework for Sustainment & Maintenance! 🎯🔧 Check out how our innovation is shaping the future of digital sustainment: 🔗 arminstitute.org/news/digita… #AI #Robotics #ManufacturingInnovation
2
1
3
502
A good hand can push intelligence development. Introducing Eyesight Hand, equipped with full-hand high-res tactile sensors and proprioceptive actuators. It is compliant, agile, and powerful. Good tactile sensing makes learning more efficient and robust. Shout out to Branden!
7
45
3
225
23,616
Stay tuned!
[1/4] 🌟Sneak Peek: SPARK in Action! 🦾 Previewing Safe Protective & Assistive Robot Kit (SPARK)—a modular toolbox designed to enhance safety in humanoid autonomy and teleoperation. Safety isn't just a feature—it's the foundation for humanoids to truly integrate into human life. SPARK filters risky actions, ensuring humanoids can achieve their objectives securely across tasks—from lab experiments to real-world deployments. With SPARK, you can innovate fearlessly, knowing safety is always guaranteed. Powered by Safe Set Algorithm (SSA), SPARK is built to: ✅ Configure safety behaviors with ease ⚖️ Balance safety and performance 🤖 Integrate with Unitree G1 + Apple Vision Pro 🔧 Support customization for other systems Stay tuned for the full release in a few weeks 🚀 Please see our website for the paper and more details! 🌐 Website: intelligent-control-lab.gith… @ICL_at_CMU @CMU_Robotics @CarnegieMellon @UnitreeRobotics #Robotics #HumanoidSafety #AIInnovation
2
13
2,535
#CoRL2024 accepted!🌈 Our work KOROL developed a linear dynamics model using object features that capture key information for robotic manipulation, outperforming models that rely on GT object states. Code: github.com/hychen-naza/KOROL
1
13
3
33
5,000
How can we solve MPC with neural dynamics efficiently but still with safety guarantees? Check out our #L4DC2024 poster "Real-Time Safe Control of Neural Network Dynamic Models with Sound Approximation" with Prof. Jianglin Lan and Prof. Changliu Liu @ChangliuL @ICL_at_CMU
1
1
7
609
Annoyed by trying out control designs one by one for each new task? Just ask Meta-Control!
🤔 How can a robot accomplish heterogeneous tasks, with precision and guarantees? 🚀 Introducing Meta-Control: Automating model-based control synthesis for heterogeneous skills with LLM! meta-control-paper.github.io…
1
5
616
Center-of-mass regulation + feet/hands on the G1 humanoid.
Differential inverse kinematics controller on a Go1 quadruped regulating the pose of the root body and the position of the feet.
4
17
1
185
30,079
Model-Free Reinforcement Learning (MFRL) has been alluring, especially with supercharged compute with physics on GPU. However, the methods use 0-th order gradients, and are often not the best optimizers. Can we do better than PPO in continuous control for robotics? Turns out yes! 🥳 tl;dr: Faster, better RL than PPO in continuous control 💪 adaptive-horizon-actor-criti… The answer lies in using more information from the simulation. We are juicing the simulation on GPU as it is, why not use it for gradients as well? This has been a driving question in a series of our works. We first studied this problem in ICLR 2022 paper on Short Horizon Actor Critic short-horizon-actor-critic.g… Naive gradient based methods are stuck in local minima and have exploding/vanishing gradients. SHAC solved this problem truncated rollouts and model based value estimation, where the model is Differentiable Sim. This boosted sample efficiency and wall-clock time immensely especially in high dimensional systems such as humanoids Yet, given enough compute PPO often caught up. Our follow up paper on on Adaptive Horizon Actor Critic at ICML 2024 discovers the cause and provides a fix. adaptive-horizon-actor-criti… However, we find that even when given ground-truth dynamics, not all gradients are useful due to sample error. 1st-Order Model-Based Reinforcement Learning methods employing differentiable simulation provide gradients with reduced variance but are susceptible to bias in scenarios involving stiff dynamics, such as physical contact. We find that back-propagating through contact and long trajectories drastically reduces gradient accuracy. Using this insight, we propose AHAC to dynamically adapt its roll-out horizon to avoid differentiating through stiff contact. AHAC is a first-order model-based RL algorithm that learns high-dimensional tasks in minutes (wall clock) and outperforms PPO by 40%, even in the limit of data provided to PPO. This work is led by @imgeorgiev alongside @krishpopdesu, @xujie7979, @eric_heiden and ample assistance from warp team at @NVIDIARobotics (@milesmacklin)
8
68
3
395
52,356