LLM-mediated Dynamic Plan Generation with a Multi-Agent Approach
This work addresses the need for versatile planning methods in robotics, autonomous vehicles, and smart systems, though it appears incremental as it builds on existing LLM and multi-agent approaches.
The researchers tackled the problem of generating adaptable planning networks for dynamic environments by using GPT-4o to automatically create multi-agent networks based on environmental status and goals, confirming that these networks showed higher generality compared to manually constructed ones.
Planning methods with high adaptability to dynamic environments are crucial for the development of autonomous and versatile robots. We propose a method for leveraging a large language model (GPT-4o) to automatically generate networks capable of adapting to dynamic environments. The proposed method collects environmental "status," representing conditions and goals, and uses them to generate agents. These agents are interconnected on the basis of specific conditions, resulting in networks that combine flexibility and generality. We conducted evaluation experiments to compare the networks automatically generated with the proposed method with manually constructed ones, confirming the comprehensiveness of the proposed method's networks and their higher generality. This research marks a significant advancement toward the development of versatile planning methods applicable to robotics, autonomous vehicles, smart systems, and other complex environments.