約50ms の低遅延で高い応答性を実現する 人型ロボット遠隔操作システム (全身制御フレームワーク) youtu.be/9Qb57bzvzO4 #bipedal #humanoidrobot #teleoperation #RealTime #LowLatency #ControlSystem #ExtremControl

Whole-body control framework that reduces end-to-end latency of teleoperation to 50 ms. [📍it’s open source] Teleoperating a humanoid usually feels like driving through lag. This brings it close to real-time. A new control framework called ExtremControl cuts latency to about 50 ms. Older systems sit around ~200 ms, which is enough delay to break balance and fine manipulation. Instead of translating the entire human body into robot joints, it directly maps hands and feet in space. That tiny change removes most of the processing overhead. They also predict motion velocity, not only position. So the robot starts moving where you’re going, not where you were. Result: movements feel continuous instead of delayed And it works with any policy or teleop stack because the control layer is independent. If you care about dexterity, data collection, or safe remote operation, latency is often the real bottleneck. Reducing it changes what tasks are even possible. Thanks for sharing, @ziyanx02! 📍Code: github.com/UMass-Embodied-AG… Project: owenowl.github.io/extremcont… —— if it matters in AI or Robotics you'll read it here first: 22astronauts.com

ExtremControl Low-Latency Humanoid Teleoperation with Direct Extremity Control 既存のヒューマノイド・テレオペは、人→ロボの全身リターゲティング（IK/最適化）や位置PD中心の設計が多く、結果として反応がモタつきやすい 多くの既存システムは end-to-end で ~200ms 前後の遅延が見られる 全身IK/最適化を毎回解くのをやめて、手足など末端リンクの姿勢を直接コントロールし計算軽量化 さらに位置PDだけでなく目標位置の差分から作った目標速度をフィードフォワードで入れて、低レベル応答を速くする 結果、end-to-end遅延を最小≈50ms級（表だと≈54ms）まで短縮 身体動作を取るためのハード（SteamVR / OptiTrack等）が要る owenowl.github.io/extremcont… github.com/UMass-Embodied-AG…

@ziyanx02 is one of the best talents on humanoid control that I know 😉 ExtremControl will be the new standard for the humanoid teleportation more applications!

2) While most humanoid teleoperation systems focus on training a general motion tracker and reusing motion preprocessing pipelines, it’s time for us to revisit the overall system design. ExtremControl features with: 1. Direct Extremity Control 2. Velocity Feedforward Control

I'm excited to introduce ExtremControl — a whole-body control framework that reduces end-to-end latency of teleoperation to 50 ms (just 2.5 control cycles), breaking past the ~200 ms barrier seen in prior systems. Website: owenowl.github.io/extremcont… Code: github.com/UMass-Embodied-AG…