This research introduces a novel affordance-driven framework that enables robots to better understand their environment using Large Language Models.

• Leverages LLMs to help robots interpret affordances - the potential uses of objects in their surroundings
• Creates a symbol network approach that translates implicit human common sense into actionable robot decisions
• Provides a method for automatic affordance acquisition that bridges the gap between human understanding and robot capabilities

This engineering breakthrough could dramatically improve human-robot interaction by allowing robots to make more intuitive decisions in dynamic environments without explicit programming.

Proposition of Affordance-Driven Environment Recognition Framework Using Symbol Networks in Large Language Models