Kepler v0.1 trains robot vision, touch, and pose into one shared world model
freelerobot · x · 2026-07-23
- Menlo Research introduces Kepler v0.1, an open research experiment where a robot’s vision, body position, and touch are trained together.
- The core idea is that the senses can predict one another: from camera input alone, the robot can infer some information that previously required touch.
- One encoder works across four different robots, and disagreement between sensory streams becomes a safety signal.
- The team says the project is being trained in public, with failures published alongside results and the whole pipeline shared openly.
More from Embodied
- Tesla says Robotaxi unsupervised miles are growing over 10% week over week — davidpattersonx · 2026-07-23
- Tesla says it is converting Fremont Model S and X lines to build Optimus — CyberRobooo · 2026-07-23
- Cleaning robot took six years to reach real homes, and the team is already planning taller robots — Scobleizer · 2026-07-23
- A humanoid robot walking-gait simulator is now in progress — i_bioloid · 2026-07-23
- Tesla says robotaxi launches are trending toward “entire states” at once — mitchdeg · 2026-07-23
- A robot finger actuation demo answers the question with a simple yes — IanPritchard · 2026-07-23