Large language models can be used to teach robots how to tidy up according the user’s preferences
Different people tend to have unique needs and preferences–particularly when it comes to cleaning or tidying up. Robots that are designed to assist humans in housework, such as robots for home cleaning, should be able, ideally, to perform tasks according to their individual preferences.
Researchers from Princeton University and Stanford University have recently been working to personalize home robots’ assistance using large language model (LLMs), an artificial intelligence class that has become increasingly popular since the release of ChatGPT. The researchers’ approach, which was presented in a pre-published paper on arXiv and tested initially on a mobile robot named TidyBot, designed to clean up indoor environments, was first tested.
In their paper, Jimmy Wu, Rika Anthonyova, and their co-authors wrote that for a robot personalize physical assistance, it needs to learn user preferences which can be applied to other scenarios. In this paper, we examine personalization of housecleaning with robots capable of tidying up rooms by picking objects up and putting away.
Source:
https://techxplore.com/news/2023-06-robots-tidy-based-user-large.html
