Figure's dancing robots spark debate: is tracker-based RL the wrong strategy?
KyleMorgenstein · x · 2026-11-20
Chris Paxton reiterates that he still thinks tracker-based RL is the wrong strategy for robotics. Responding to an interview quote claiming Chinese dancing robots use a "180-degree different" neural architecture, Carlos Ponx points out the irony: Figure itself uses a SONIC-style planner-tracker architecture, and that tracker research lineage traces back to the dancing robots. A notable technical debate over humanoid RL approaches (planner-tracker vs end-to-end).
More from Embodied
- Fully driverless Tesla Cybercab with Starlink spotted in Dallas ahead of ride launch — mitchdeg · 2026-11-20
- Analyst: Robot hardware demand will overwhelm supply by orders of magnitude for years — Rewkang · 2026-11-20
- Teamily launches Video Agent letting AI agents see through your AI glasses — CurieuxExplorer · 2026-11-20
- Verne Robotics' 8.2B open-weight V0.5 hits 99.3% on LIBERO, dubbed the Thinking Machines of robotics — chris_j_paxton · 2026-11-20
- Legged-robot state estimator from proprioception alone, leveraging contact events — chris_j_paxton · 2026-11-20
- 60-Day Physical AI Challenge: Simulating a Robot Arm in MuJoCo, No Hardware — DynamicWebPaige · 2026-11-20