LLM-Flock: LLM-Driven Decentralized Multi-Robot Flocking via Influence-Based Consensus with Collision Avoidance
LLM-Flock is a decentralized system that allows multiple robots to form a coordinated group using Large Language Models (LLMs) for planning. Robots first generate individual formation plans, then use an influence-based consensus protocol to align their plans without central control. This method improves stability and convergence compared to previous LLM approaches, proving LLMs can be used for complex multi-robot coordination.