Wang Xiao (王啸)Ph.D. Student
Department of Computer Science and Engineering |
 |
Revolutionizing Therapy with LLM-Robot Agents
Abstract
This project explores the integration of Large Language Models (LLM) with the Misty robot, aiming to enhance robotic interaction capabilities through advanced natural language processing techniques. By interfacing LLMs with Misty's control system, the robot gains the ability to comprehend and generate natural language, facilitating more intuitive and flexible human-robot interactions. The core development includes an interface that allows Misty to parse verbal inputs into actionable commands and execute complex tasks, ranging from understanding human directives to engaging in dialogue and task planning with natural language feedback. Utilizing state-of-the-art LLMs, such as those from OpenAI's GPT series, ensures high accuracy in language understanding and generation, while also focusing on optimizing the efficiency and response time for real-time interactions. The results from experiments and user testing highlight the significant potential of leveraging LLMs to improve interaction quality and task efficiency with Misty robots, offering a powerful new capability for robotics and paving the way for future innovations in robot technology and natural language processing applications.
|
Last Updated on March, 2024
|
