Amazon FAR, CMU and Stanford's FIBER trains Unitree G1 to chain skills for delivery runs
CyberRobooo · x · 2026-10-24
FIBER, a research project involving Amazon FAR, Carnegie Mellon and Stanford, uses a Unitree G1 to show how humanoids can chain discrete skills: catching a box thrown to a random spot, grabbing boxes from different positions, shouldering open a heavy door, and crouching under a low beam while carrying a box — the skill set a delivery humanoid would need. The approach trains separate skills, merges them into a shared behavior model, and runs a high-level planner at 5 Hz atop a 50 Hz low-level controller, trained in simulation and tested on a real G1. Still early: 42% average success across three simulated tasks, no end-to-end delivery yet.
More from Embodied
- Tesla starts production of Optimus humanoid robots, targets 20,000 units per week — LexiLove · 2026-10-24
- iPhone Duo in stock for immediate pickup at French Apple Stores, from €3,089 — mitchdeg · 2026-10-24
- PWAs can't use foldable screens on the Duo yet, but two Chrome flags unlock the APIs — mitsuhiko · 2026-10-24
- Robot ducks trained in RTX Spark simulation flip live on stage at NVIDIA event — NandoDF · 2026-10-24
- Baidu's Apollo Go and Lyft's Freenow unveil RT6 robotaxi in London, targeting 2027 driverless launch — Baidu_Inc · 2026-10-24
- $3000 folding iPhone hit by bright screen lines, hinge design fault suspected — LinusEkenstam · 2026-10-24