Communications of the ACM
Cleaning Up with TidyBot
In testing, TidyBot correctly determined the category of objects and where to put them—as in, place toys in the box and put clothes on floor—for 85% of objects.
Credit: Princeton Engineering News
A multi-institutional team of engineers mated a mobile robotic arm to a vision model and a large language model to create the TidyBot cleaning robot.
The robot's arm is attached to a wheeled base; its pincer hand can open drawers and retrieve objects, enabling it to carefully put dishes in the sink or toss clothes onto a laundry pile.
TidyBot can differentiate between types of objects with its combined camera/vision model, and the large language model with which it is programmed allows it to learn how carefully or casually it can handle different items.
The robot was 85% accurate in categorizing objects and where to put them.
From Princeton Engineering News
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found