This research introduces V2V-LLM, a novel approach using multi-modal large language models to enable cooperative autonomous driving through vehicle-to-vehicle communication.

• Addresses critical sensor reliability challenges by allowing vehicles to share perception data
• Utilizes large language models to interpret multi-modal inputs from multiple vehicles
• Demonstrates significant improvements in driving safety and decision-making quality
• Creates a comprehensive framework for cooperative planning, not just perception

This breakthrough matters for security by enhancing system resilience against sensor failures and occlusions, creating redundancy that substantially improves autonomous driving safety in real-world conditions.

V2V-LLM: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multi-Modal Large Language Models