Astribot CoMind lets robots share views, lifting occluded-pick accuracy to 88.9%
Scobleizer · x · 2026-11-14
Astribot unveiled CoMind – Be My Eye, a framework for its S1 robots to work together by sharing views.
- Each robot learns which visual details its partner needs, compresses and sends them; the partner fuses that input with its own observations to guide actions.
- Demoed: two S1s restocking a fridge, assembling a box, folding a blanket together.
- In an occlusion test, a robot that couldn't see which of three boxes held a blue plate chose correctly 88.9% of the time with CoMind, up from 38.9% (18 trials per condition).
- Scobleizer notes implications for autonomous cars and robot collaboration — robots can effectively "see around corners."
More from Embodied
- LabMate robot demo: Gemma 4 on a reBot arm fetches missing lab tools — agihouse_org · 2026-11-14
- Gemma hackathon winner: robot brain failover to another machine in 625ms, no cloud — agihouse_org · 2026-11-14
- Cybertruck police car spotted at Starbase with Starship in the background — shivon · 2026-11-14
- PiHand: latent hand-object interaction priors cut dexterous-hand RL training data 10x — KyleMorgenstein · 2026-11-14
- OpenAI insider: robotics is "alive and well", with 15 open roles hiring — ashwinl · 2026-11-14
- Teleoperating Mundane's robot via force-feedback exoskeleton: can feel empty vs full bottle — chris_j_paxton · 2026-11-14