Berkeley's S1 robot shows self-correction: realigns a misattached wheel on its own
thisdudelikesAI · x · 2026-09-10
Berkeley professor Deepak Pathak demoed his S1 robot's error recovery: when a wheel attached via video prompt didn't align properly during deployment, S1 noticed and realigned it itself. Pathak explains that S1 watches the prompt, understands the goal, and improvises to achieve it — "you learn a lot more about AI when it screws up a task."
Related event: Skild S1 Robot Self-Corrects After Misaligning Wheel(4 posts)→
More from Embodied
- StackChan robot performs an AI-generated skit, orchestrated by Hermes Agent via JSON script — alexcovo_eth · 2026-10-27
- DONIC hits 86.79% on bimanual dexterous tracking, crushing prior SOTA's 9.43% — liuziwei7 · 2026-10-27
- Bedrock Robotics raises $600M as its fully autonomous excavators do paid work on live job sites — Sethwinterroth · 2026-10-27
- AW-0 hits 80.8% on RoboTwin tasks, beating strongest baseline by 16 points — trevordarrell · 2026-10-27
- AW-0's code stream fixes world model prediction: picks mango where baseline grabs apple — trevordarrell · 2026-10-27
- AW-0 runs at 230ms on a single RTX 5090, 2.4x faster than autoregressive code planning — trevordarrell · 2026-10-27