Generation of Indoor Open Street Maps for Robot Navigation from CAD Files

Abstract

The deployment of autonomous mobile robots is predicated on the availability of environmental maps, yet conventional generation via SLAM (Simultaneous Localization and Mapping) suffers from significant limitations in time, labor, and robustness, particularly in dynamic, large-scale indoor environments where map obsolescence can lead to critical localization failures. To address these challenges, this paper presents a complete and automated system for converting architectural Computer-Aided Design (CAD) files into a hierarchical topometric OpenStreetMap (OSM) representation, tailored for robust life-long robot navigation. Our core methodology involves a multi-stage pipeline that first isolates key structural layers from the raw CAD data and then employs an AreaGraph-based topological segmentation to partition the building layout into a hierarchical graph of navigable spaces. This process yields a comprehensive and semantically rich map, further enhanced by automatically associating textual labels from the CAD source and cohesively merging multiple building floors into a unified, topologically-correct model. By leveraging the permanent structural information inherent in CAD files, our system circumvents the inefficiencies and fragility of SLAM, offering a practical and scalable solution for deploying robots in complex indoor spaces. The software is encapsulated within an intuitive Graphical User Interface (GUI) to facilitate practical use. The code and dataset are available at https://github.com/jiajiezhang7/osmAG-from-cad.

Figures (6)

• Figure 1: The automatic generation pipeline showcasing key outputs. (\ref{['fig:teaser1']}) The original architectural CAD drawing, containing numerous non-essential layers and elements. (\ref{['fig:teaser2']}) The result after topological segmentation into an Area Graph, which initially exhibits over-segmented polygons. (\ref{['fig:teaser3']}) The refined Area Graph exported to the OpenStreetMap format and visualized in JOSMosm_wiki:JOSM; this stage includes a post-processing step where small, insignificant areas are pruned or merged. (\ref{['fig:teaser4']}) A visualization of our enhanced OSM rendered by OpenIndoorosm_wiki:openindoor, demonstrating that our generated map is fully compatible with standard OSM tools.
• Figure 2: An overview of the proposed map generation pipeline. The system takes a raw CAD floor plan as input and sequentially processes it through stages of preprocessing, topological segmentation, refinement, serialization, and semantic attribution, culminating in a multi-story, enhanced OSM map ready for robotic navigation.
• Figure 3: Illustration of the AreaGraph topological structure derived from a floor plan. Nodes correspond to segmented polygonal areas (e.g., rooms, corridors), and edges represent the passages connecting them, forming the foundational topological map for navigation.
• Figure 4: Results of Hierarchical Multi-Floor Map Fusion. Figures \ref{['fig:merge1']},\ref{['fig:merge2']},\ref{['fig:merge3']} depict three distinct hierarchical levels of a university building as rendered in the OpenIndoor viewerosm_wiki:openindoor. Figure \ref{['fig:merge4']} showcases an enhanced OSM map visualized in Rviz after being processed by osmAG parser feng2023osmag. The red line segments represent vertical passages (e.g. elevators and stairs) that topologically connect adjacent floors.
• Figure 5: Qualitative results across 5 diverse architectural styles. From top to bottom: (1) Original CAD; (2) Intermediate AreaGraph; (3) Refined OSM in JOSM; (4) Rendered 3D View. The high structural fidelity across styles confirms the method's generalization capability.