I've long wondered if we can make a humanoid robot do a 𝘄𝗮𝗹𝗹𝗳𝗹𝗶𝗽 - and we just made it happen by leveraging 𝗢𝗺𝗻𝗶𝗥𝗲𝘁𝗮𝗿𝗴𝗲𝘁 with BeyondMimic tracking! This came after our original OmniRetarget experiments, with only minor tweaks to RL training: relaxing a

Humanoid motion tracking performance is greatly determined by retargeting quality! Introducing 𝗢𝗺𝗻𝗶𝗥𝗲𝘁𝗮𝗿𝗴𝗲𝘁🎯, generating high-quality interaction-preserving data from human motions for learning complex humanoid skills with 𝗺𝗶𝗻𝗶𝗺𝗮𝗹 RL: - 5 rewards, - 4 DR

Excited to organize Workshop on Learning Meets Model-Based Methods for Contact-Rich Manipulation @ ICRA 2025! We welcome submissions on a range of topics—check out our website for details: https://t.co/edgPLeJDId Join us for an incredible lineup of speakers! #ICRA2025

While neural networks have made a seismic impact on the design of robotic controllers, it’s been difficult to verify the safety & stability of complex robots powered by them 🧵 Researchers from MIT CSAIL & elsewhere have developed approaches to rigorously certify Lyapunov

The next honorable mention for the 2023 IEEE Transactions on #Robotics King-Sun Fu Memorial Best Paper Award is a paper by researchers @MIT_CSAIL that enables global planning for contact rich system using local smoothing. https://t.co/z90DsDlvAZ #SmoothingMethods

Introducing a “nonlinear equivalent” of LQR controllers! ⚙️ CSAIL researchers synthesize approximate optimal controllers for nonlinear dynamical systems w/optimality & stability guarantees. To achieve this, their work finds both under-and over-approximations of the value