LanguageMPC: Large Language Models as Decision Makers for Autonomous Driving

TL;DR

LanguageMPC leverages large language models for autonomous driving decision-making, enhancing reasoning, generalization, and interpretability in complex scenarios through novel algorithms and integration with low-level controllers.

Contribution

This work introduces a novel framework using LLMs as decision-makers in autonomous driving, enabling high-level reasoning and seamless integration with control systems.

Findings

LLM-based decisions outperform baseline methods in single-vehicle tasks

The approach effectively handles complex driving behaviors and multi-vehicle coordination

LLMs provide human-like commonsense reasoning for autonomous driving scenarios

Abstract

Existing learning-based autonomous driving (AD) systems face challenges in comprehending high-level information, generalizing to rare events, and providing interpretability. To address these problems, this work employs Large Language Models (LLMs) as a decision-making component for complex AD scenarios that require human commonsense understanding. We devise cognitive pathways to enable comprehensive reasoning with LLMs, and develop algorithms for translating LLM decisions into actionable driving commands. Through this approach, LLM decisions are seamlessly integrated with low-level controllers by guided parameter matrix adaptation. Extensive experiments demonstrate that our proposed method not only consistently surpasses baseline approaches in single-vehicle tasks, but also helps handle complex driving behaviors even multi-vehicle coordination, thanks to the commonsense reasoning capabilities of LLMs. This paper presents an initial step toward leveraging LLMs as effective decision-makers for intricate AD scenarios in terms of safety, efficiency, generalizability, and interoperability. We aspire for it to serve as inspiration for future research in this field. Project page: https://sites.google.com/view/llm-mpc