Description

The integration of context-aware large language models (LLMs) into social robotics has the potential to transform human-robot interaction by enabling robots to communicate with greater intelligence, empathy, and adaptability. This research investigates how LLMs can enhance a robot's ability to understand and respond to user intent, emotional states, and situational dynamics, fostering more engaging and personalized interactions. By designing and implementing a context-aware framework, the study explores key challenges, including real-time language processing, ethical implications, and maintaining user privacy. Experiments conducted in simulated and real-world scenarios demonstrate that LLM-enhanced social robots significantly outperform traditional systems in contextual understanding, conversational fluency, and user satisfaction. These findings underscore the promise of LLMs in advancing social robotics and lay the groundwork for future applications in human-centered AI.

More info: https://upcommons.upc.edu/handle/2117/414482