Kyle Morgenstein: full joint space is the wrong abstraction for high-level robot agents
KyleMorgenstein · x · 2026-11-20
Responding to Chen Tessler's pushback that tracking/human motion priors generalize well, Kyle Morgenstein argues:
- Full joint space is the wrong abstraction for higher-level agents to plan in
- There are great priors to mine from human motion data, but he isn't sold that naive joint tracking is the right interface for humanoid control
Related event: Debate: joint-space planning vs human motion priors in robot control(2 posts)→
More from Embodied
- Roboticist argues full joint space is the wrong abstraction for high-level agents — KyleMorgenstein · 2026-11-20
- US Physical AI Runs on Private CAPEX, Frontier Models, and Energy Flexibility — ingliguori · 2026-11-20
- Skild AI open-sources code, ending speculation about its robotics work at CoRL — KyleMorgenstein · 2026-11-20
- Skild AI releases Steel 0.2 robot simulation platform with 81 environments and magnetic contacts — KyleMorgenstein · 2026-11-20
- STASIS (YC F26) launches smart glasses with 6 sensors tracking eye strain at 92% accuracy — LinusEkenstam · 2026-11-20
- NVIDIA's GR00T 2 Controls Real Dexterous Hands End to End, Fully Onboard — rsasaki0109 · 2026-11-20