Stanford team teaches an AI agent a pool table's physics by letting it take real shots (CoRL)
drmapavone · x · 2026-11-09
A Stanford AI Lab team presented a late-breaking contribution at CoRL: an agent takes real shots on a physical pool table, observes outcomes, and iteratively improves its simulation of the table's dynamics.
The thesis: closing the visual sim-to-real gap is one thing, but the dynamics gap is much harder — rather than just observing the world, agents should actively experiment with it and learn physics through autonomous interaction. The team frames this as a key capability for Physical AI, with collaborators including Xilun Zhang, Matthew Foutter, and Luis Pabon.
More from Embodied
- iPad mini 8 hands-on: 1g heavier and a third the price of iPhone Air, 60Hz capped — oran_ge · 2026-11-09
- Dooder puts an AI with on-device hearing and voice entirely on the Apple Watch — breagz · 2026-11-09
- Furong Huang Lab brings six human-video-to-robot works to CoRL 2026 — furongh · 2026-11-09
- 3D animated teardown explains how the open-source local-AI microduck robot works — Individual-Dot5488 · 2026-11-09
- Three CMU robot papers spotlighted at CoRL 2026: memory, vision-touch steering, world models — pabbeel · 2026-11-09
- CoRL 2026 balloons to 5,000+ attendees; Princeton PRISM brings 6 robotics papers including oral MolmoB0T — shahdhruv_ · 2026-11-09