NVIDIA & Berkeley open source tactile robotics method T-Rex
DrJimFan · x · 2026-08-22
NVIDIA and UC Berkeley open-sourced T-Rex, a tactile learning methodology for robotics addressing the lack of touch sensing. Key contributions include: (1) Architecture: A mixture-of-transformer running async clocks—a slow visuomotor expert plans motion while a fast tactile expert refines it at 4 touch ticks per vision tick. (2) Dataset: The largest tactile dataset released (50 hours, 5.5k episodes) collected on SOTA 22-DOF hardware. (3) Training: Extends EgoScale, using egocentric video for pretraining and tactile robot play for mid-training, bridging the gap to contact-rich manipulation. Tactile sensing is emphasized as critical for the "last mile" of manipulation.
Related event: NVIDIA and UC Berkeley Open-Source T-Rex Tactile Robot Learning(2 posts)→
More from Embodied
- Expert turns a 'useless' $900 robot vacuum into a cleaning beast in 15 mins — Aiden_Tech_Ai · 2026-08-22
- Oura faces class-action lawsuit alleging 'coin flip' accuracy for sleep stages — Polymarket · 2026-08-22
- Wormable RCE Vulnerabilities Found in Unitree Robots — matthew_d_green · 2026-08-22
- Eve launches '3D Chat Room' beta on TestFlight — Kuprel · 2026-08-22
- τ0-VLA: Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Search — sii-research · 2026-08-22
- Robotics in 2027: One-Shot Learning and Exponential Growth — LinusEkenstam · 2026-08-22