Unitree G1 post-trained with Gemini Robotics-ER 1.5 for real-time physical hazard intervention
DynamicWebPaige · x · 2026-11-10
Shared by Google's Dynamic Web Paige: one team post-trained Gemini Robotics-ER 1.5 (SFT + RL) for "safety-thinking", enabling a Unitree G1 humanoid to detect physical hazards (e.g., a toddler swallowing a pill) and intervene in real time. A second team uses Nano Banana image generation as "visual imagination" for planning—synthesizing goal images for deformable object manipulation from a scene plus prompt.
More from Embodied
- Amazon Ring unveils $250 autonomous home security drone that patrols and streams live video — Polymarket · 2026-11-10
- NVIDIA-led study: unblocking VLM gradients in VLA training lifts manipulation progress from 10.3% to 32.9% — stepjamUK · 2026-11-10
- Google DeepMind launches Gemini Robotics 2.5 with 96% accuracy sorting Mercedes-Benz parts — GoogleDeepMind · 2026-11-10
- SlashVibe is a physical desk controller for managing multiple running AI coding agents — Scobleizer · 2026-11-10
- Path Robotics raises $300M Series E at $2.45B valuation to expand physical AI — Rewkang · 2026-11-10
- Unitree Go2 robot asks clarifying questions via Gemini and Bayesian active learning — DynamicWebPaige · 2026-11-10