BIM-to-Robot Mapping: Constructing IFC-Referenced Occupancy Grids and Semantic Metadata for Rule-Based Navigation

Abstract

Aiming to solve the problems of insufficient BIM semantic fusion, poor dynamic interference robustness, and missing path compliance in existing robot navigation systems, this paper proposes a navigation framework based on BIM semantic enhancement. We construct an OGM map and a semantic mapping dictionary by extracting the geometric topology and functional partitions from the BIM model. A hierarchical localization strategy is designed, which uses semantically-constrained particle filtering for global localization and Cartographer subgraph matching to suppress dynamic interference for local pose tracking. Combined with spatial functional labels in the semantic dictionary, elements such as fire zoning and prohibited areas are encoded as path costs by BIM semantic constraints, ensuring that navigation paths meet building regulations. This framework effectively maps BIM semantics to robot navigation, providing a scalable solution for autonomous navigation in complex building scenarios.