Hang Liu
@uint8_Lau
@UMich 🤖Robotics |🧠 Learning |🦿 Legged Robot
🛹Can robots perform cool skateboarding tricks?! 🥰SURE!! 💫Discrete-Time Hybrid Automata Learning: Legged Locomotion Meets Skateboarding Paper: arxiv.org/abs/2503.01842 Website:umich-curly.github.io/DHAL/ Enjoy our video! We hope it brings you joy!
Happy 2025! May the robots finally get smarter and start helping humans. Robot trained to skateboard by @uint8_Lau and @SangliTeng
🤖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
Meet 𝐀𝐌𝐎 — our universal whole‑body controller that unleashes the 𝐟𝐮𝐥𝐥 kinematic workspace of humanoid robots to the physical world. AMO is a single policy trained with RL + Hybrid Mocap & Trajectory‑Opt. Accepted to #RSS2025. Try our open models & more 👉…
Command humanoids *directly* with natural language? Introducing LangWBC, a generative, end-to-end policy that turns natural language into real-world whole-body humanoid control! 💬→🦿Smooth, robust, surprisingly intuitive! See more 👉 LangWBC.github.io #RSS2025
Introducing OceanSim: A High-Fidelity, GPU-Accelerated Underwater Robotics Simulator 🌊 Explore our project website: umfieldrobotics.github.io/OceanSim/ OceanSim is built for realistic & efficient underwater robot simulation. #Robotics #UnderwaterRobotics #OpenSource
1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL) can significantly boost performance, ranging from doubling performance to 50x on a diverse suite of robotic tasks. Webpage+Paper+Code: wang-kevin3290.github.io/scaling-crl/
Atlas is demonstrating reinforcement learning policies developed using a motion capture suit. This demonstration was developed in partnership with Boston Dynamics and @rai_inst.
RL is notoriously sample inefficient. How can we scale RL on tasks much slower to simulate than rigid body physics, such as soft bodies? In our #ICLR2025 spotlight, we introduce both a new first-order RL algorithm, SAPO, and differentiable simulation platform, Rewarped. 1/n