COMRES-VLM: Coordinated Multi-Robot Exploration and Search using Vision Language Models

Researchers developed COMRES-VLM, a new framework using Vision Language Models to coordinate multiple robots for exploration and object search in indoor environments. The system achieved 10.2% faster exploration and 55.7% higher search efficiency compared to existing methods, while enabling natural language-based human guidance.

• →COMRES-VLM framework uses Vision Language Models to coordinate multi-robot systems for autonomous exploration and object search.
• →The system outperformed state-of-the-art methods with 10.2% faster exploration completion and 55.7% higher object search efficiency.
• →Testing involved up to six robots in large-scale simulated indoor environments with real-time coordination.
• →The framework enables natural language-based object search, allowing human operators to provide semantic guidance.
• →COMRES-VLM integrates frontier cluster extraction and topological analysis with VLM reasoning for globally consistent waypoint assignments.