One policy for millions of robots: team builds generalist motor control policy trained on 200+ real robot models
breadli428 · x · 2026-10-02
Researcher Bo Ai detailed the team's foundation motor control policy: a single generalist policy that can control both ground robots (legged) and aerial robots (drones), showing exciting transfer to unseen robot embodiments.
Key facts:
- Trained across millions of embodiment variations for multi-embodiment generalist learning
- Collected 200+ real-world robot models to shape the embodiment distribution
- Over billions of cumulative simulation steps
- The team is recruiting collaborators to contribute robot models, incorporate them into training, and deploy on real hardware
breadli428 amplified the work by Nico Bohlinger and Bo Ai on multi-embodiment generalist learning.
More from Embodied
- Kaigen dev confirms its AI 3D demos run entirely inside the Unity engine — gdechichi · 2026-10-02
- Cartwheel teases kinematics data breakthrough with a wink — andrew_n_carr · 2026-10-02
- Teleop a humanoid just by wearing glasses, and it sees through them too — danfei_xu · 2026-10-02
- Clone Robotics shows off Torso 3 humanoid hand, betting on an unconventional design path — MarwaEldiwiny · 2026-10-02
- Developer shows AI wearable dot proactively pinging him about schedule choices — dkundel · 2026-10-02
- Figure sells limited-edition artifacts made from melted-down F.02 robots — adcock_brett · 2026-10-02