SignLoc: Robust Localization using Navigation Signs and Public Maps

2025-08-25

AI Summary

Robots operating in human environments often fail to effectively leverage navigation signs and publicly available maps (e.g., floor plans, OpenStreetMap) for global localization. Method: This paper proposes a cross-modal global localization framework that operates without prior map construction. It extracts the topological structure from public maps and integrates directional and positional semantics of navigation signs into a probabilistic observation model within a Monte Carlo localization framework, enabling robust semantic–topological alignment. Contribution/Results: To our knowledge, this is the first approach jointly modeling text recognition, multi-sensor measurements, and public map topology. It achieves initialization and high-precision localization using only 1–2 observed navigation signs. Evaluated across large-scale real-world environments—including campuses, shopping malls, and hospitals—the method significantly outperforms conventional approaches in both localization success rate and accuracy.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Security and Privacy: Large-scale security measurements

Abstract

Navigation signs and maps, such as floor plans and street maps, are widely available and serve as ubiquitous aids for way-finding in human environments. Yet, they are rarely used by robot systems. This paper presents SignLoc, a global localization method that leverages navigation signs to localize the robot on publicly available maps -- specifically floor plans and OpenStreetMap (OSM) graphs--without prior sensor-based mapping. SignLoc first extracts a navigation graph from the input map. It then employs a probabilistic observation model to match directional and locational cues from the detected signs to the graph, enabling robust topo-semantic localization within a Monte Carlo framework. We evaluated SignLoc in diverse large-scale environments: part of a university campus, a shopping mall, and a hospital complex. Experimental results show that SignLoc reliably localizes the robot after observing only one to two signs.

Problem

• Localizing robots using navigation signs and public maps
• Matching detected signs to map graphs without prior mapping
• Enabling robust topo-semantic localization in large-scale environments

Innovation

• Leverages navigation signs and public maps
• Extracts graph from maps without prior mapping
• Uses probabilistic model for topo-semantic localization