SD++: Enhancing Standard Definition Maps by Incorporating Road Knowledge using LLMs

TL;DR

This paper introduces SD++, a pipeline that uses large language models to enhance standard definition maps with detailed, location-specific road information from manuals, aiming to bridge the gap between SD and high-definition maps.

Contribution

The paper presents a novel end-to-end method leveraging LLMs to incorporate manual-based road knowledge into SD maps, demonstrating cross-region generalization.

Findings

SD++ effectively enriches SD maps with manual-derived road info.

The approach generalizes well across California and Japan.

Multiple methods for utilizing LLMs are compared.

Abstract

High-definition maps (HD maps) are detailed and informative maps capturing lane centerlines and road elements. Although very useful for autonomous driving, HD maps are costly to build and maintain. Furthermore, access to these high-quality maps is usually limited to the firms that build them. On the other hand, standard definition (SD) maps provide road centerlines with an accuracy of a few meters. In this paper, we explore the possibility of enhancing SD maps by incorporating information from road manuals using LLMs. We develop SD++, an end-to-end pipeline to enhance SD maps with location-dependent road information obtained from a road manual. We suggest and compare several ways of using LLMs for such a task. Furthermore, we show the generalization ability of SD++ by showing results from both California and Japan.