This research integrates Large Language Models (LLMs) directly into autonomous driving systems as decision-making components, enabling human-like reasoning for complex scenarios.

• Develops cognitive pathways that allow LLMs to comprehend high-level driving information
• Creates algorithms that translate LLM decisions into actionable driving controls
• Enhances vehicle ability to handle rare events through LLM commonsense reasoning
• Improves interpretability of autonomous driving decisions

This engineering breakthrough addresses critical safety and reliability challenges in autonomous driving by combining the contextual understanding of LLMs with traditional control systems.

LanguageMPC: Large Language Models as Decision Makers for Autonomous Driving