Recova Trains Robots to Self-Recover in Simulation, Cutting Human Intervention From 87.5% to 0%
Lianhuiq · x · 2026-10-14
Recova targets a core pain point in robot data collection: teams want more real-world robot data but lack the people to keep robots running. In a small team, every failed attempt means someone must step in, fix the scene, and restart.
The approach has three parts:
- Real2Sim2Real: agents explore and learn in digital twins, where trial and error is safer and cheaper.
- Autonomous recovery: robots learn to recover from unexpected failures instead of waiting for human resets.
- Human-on-demand learning: multiple robots run in parallel, and humans step in only when autonomous recovery falls short.
A coding agent develops recovery skills in simulation; on real robots the system monitors progress, attempts recovery when things go wrong, and asks for help when needed. Those experiences, including human corrections, feed back into training.
The loop ran across four real-robot stations. On one task, observed human intervention dropped from 87.5% to 0% over four collection rounds.
Related event: Recova teaches robots to self-recover, cutting human interventions to zero(2 posts)→
More from Embodied
- China reportedly shipped 97% of the world's humanoid robots in H1 2026 — Polymarket · 2026-10-14
- Physical Intelligence opens Pi API beta, one model adapting robots from strawberry harvesting to server assembly — svlevine · 2026-10-14
- China's Tiangong Ultra robot beat Bolt's 100m record — and exposes America's robotics supply-chain dilemma — mattsheehan88 · 2026-10-14
- GoDrift shows off realtime sim2real robot demo; 2D plane eases reaction demands — jonstephens85 · 2026-10-14
- Microsoft science president Peter Lee leaves to become Mercedes F1 team's AI advisor — lazowska · 2026-10-14
- NVIDIA releases GR00T N1.7 AppleToPlate checkpoint-15000 on Hugging Face — nvidia · 2026-10-14