CMU researchers, in collaboration with NVIDIA, present ASAP, a two-stage framework for humanoid robot agility.
It pre-trains motion policies on human data, then refines them with real-world corrections using a delta action model, which adjusts for simulation mismatches.
1k placements a day—long-horizon tasks—using vision-only—learned behaviors for very precise <1cm sheet metal insertion.
Requires intense engineering across multiple disciplines.