Scientists at Switzerland's ETH Zurich have formulated a "minimal manual guidance" model to train robots to learn how to open doors and dishwashers.
The process involves the user describing the scene and action, the system planning a complicated path, and then refining it into a minimal viable route.
The researchers wrote, "Given high-level descriptions of the robot and object, along with a task specification encoded through a sparse objective, our planner holistically discovers: how the robot should move, what forces it should exert, what limbs it should use, as well as when and where it should establish or break contact with the object."
The team split the system into object- and robot-centric categories and executed demonstrations on the four-legged ANYmal robot from ETH Zurich spin-off company ANYbotics.
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found