Tag Map: A Text-Based Map for Spatial Reasoning and Navigation with Large Language Models

TL;DR

This paper introduces an explicit text-based map for spatial reasoning that efficiently integrates with large language models, enabling robots to generate actionable plans with less memory and effective scene understanding.

Contribution

The paper presents a novel explicit text-based map that supports thousands of semantic classes and integrates seamlessly with LLMs, improving localization efficiency and scene grounding.

Findings

Text-based map localizations perform comparably to open vocabulary maps

The proposed map uses 2-4 orders of magnitude less memory

Real-robot experiments validate effective task grounding with LLMs

Abstract

Large Language Models (LLM) have emerged as a tool for robots to generate task plans using common sense reasoning. For the LLM to generate actionable plans, scene context must be provided, often through a map. Recent works have shifted from explicit maps with fixed semantic classes to implicit open vocabulary maps based on queryable embeddings capable of representing any semantic class. However, embeddings cannot directly report the scene context as they are implicit, requiring further processing for LLM integration. To address this, we propose an explicit text-based map that can represent thousands of semantic classes while easily integrating with LLMs due to their text-based nature by building upon large-scale image recognition models. We study how entities in our map can be localized and show through evaluations that our text-based map localizations perform comparably to those from open vocabulary maps while using two to four orders of magnitude less memory. Real-robot experiments demonstrate the grounding of an LLM with the text-based map to solve user tasks.