Communications of the ACM
Scientists Putting LLM Brains Inside Robot Bodies
The effort to give robots AI brains is revealing big practical challenges and ethical concerns.
Credit: Christopher Payne
Robotics researchers are using large language models (LLMs) to skirt preprogramming limits.
Computer scientists at the University of Southern California developed ProgPrompt, which involves giving an LLM prompts in the Python programming language that include a sample question and solution to help restrict its answers to the range of tasks the robot can perform.
Google researchers have developed a strategy that involves giving a list of behaviors that can be performed by the robot to the PaLM LLM, which responds to human requests to the robot in conversational language with a behavior from the list.
From Scientific American
View Full Article
Abstracts Copyright © 2024 SmithBucklin, Washington, DC, USA
No entries found