Determining spatial neighborhoods in indoor space using integrated IndoorGML and IndoorPOI data

Indoor space has been one of the focal points for geospatial research as various factors such as increasing demands for application and demand for adaptive response in emergencies have arisen. IndoorGML (Indoor Geography Markup Language) has provided a standardized method of representing the topological aspect of micro-scale environments, with its extensive specifications and flexible applicability. However, as more real-world problems and needs demand attention, suggestions to improve this standard, such as representing IndoorPOI (Indoor Points of Interest), have arisen. Hence, existing algorithms and functionalities that we use on perceiving these indoor spaces must also adapt to accommodate said improvements. In this study, we explore how to define spatial neighborhoods in indoor spaces represented by an integrated IndoorGML and IndoorPOI data. We revisit existing approaches to combine the aforementioned datasets and refine previous approaches to perform neighborhood spatial queries in 3D. We implement the proposed algorithm in three use cases using sample datasets representing a real-world structure to demonstrate its effectiveness for performing indoor spatial analysis.