Driving Everywhere with Large Language Model Policy Adaptation
This addresses the challenge of widespread deployment of autonomous vehicles by enabling adaptation to new locations, though it appears incremental as it applies existing LLM capabilities to a specific domain.
The paper tackles the problem of adapting driving behavior to new environments, customs, and laws in autonomous driving by introducing LLaDA, a tool that uses large language models to interpret local traffic rules from handbooks. It shows that LLaDA outperforms baseline planning approaches on all metrics in real-world datasets and is useful in disambiguating unexpected situations in a user study.
Adapting driving behavior to new environments, customs, and laws is a long-standing problem in autonomous driving, precluding the widespread deployment of autonomous vehicles (AVs). In this paper, we present LLaDA, a simple yet powerful tool that enables human drivers and autonomous vehicles alike to drive everywhere by adapting their tasks and motion plans to traffic rules in new locations. LLaDA achieves this by leveraging the impressive zero-shot generalizability of large language models (LLMs) in interpreting the traffic rules in the local driver handbook. Through an extensive user study, we show that LLaDA's instructions are useful in disambiguating in-the-wild unexpected situations. We also demonstrate LLaDA's ability to adapt AV motion planning policies in real-world datasets; LLaDA outperforms baseline planning approaches on all our metrics. Please check our website for more details: https://boyiliee.github.io/llada.