PhD project

Public buildings such as hotels, airports or shopping malls are becoming bigger and bigger and furthermore their internal structures are getting more complex. This phenomenon results in the issue, that persons (especially foreign people) are likely to get lost inside the building or to take circuitous routes when navigating through the building. In the area of leisure time (e.g. shopping) this might be just disturbing, but in business life this could result in losing time and money. In the area of emergency situations or rescue forced, the fastest route is even more crucial, because one second can decide between life and death. Today’s navigation solutions are becoming more mature and versatile, however they are mainly designed for outdoor areas (e.g. vehicle navigation, bicycle navigation or pedestrian navigation). That is, there is an increasing demand for advanced and sophisticated indoor navigation solutions for many different areas (Cf. [1]). One of the core problems is, where adequate data could be acquired from for such indoor navigation applications. A large‐scale data acquisition by commercial data providers (e.g. Navteq etc.) is not likely for indoor environment. That is, alternative data sources (both internal and external) are required. One example for such new, alternative data sourced is Volunteered Geographic Information (VGI), describing a collective and collaborative accumulation of spatial data through a web 2.0 platform. One of the most famous examples for VGI is OpenStreetMap (OSM), whereby both data quality and quantity steadily increases. Currently, OSM mainly contains information about the outdoor space, but there are already more than 34 million building footprints as well as additional building attributes (e.g. height, color, roof shape etc.) available. The intention of this dissertation project is the usage of VGI data from OSM for the provision of indoor navigation applications. Thereby, methodologies for the extension of OSM towards indoor environments need to be defined, so that detailed information about different floor plans and distinct rooms can be acquired and provided. Based on this data, CityGML LoD4 building models will be generated. CityGML is chosen due to the fact that it represents an internationally accepted standard for exchanging 3D city models including semantic information. These (automatically) generated building models will then be used for derivating 3D routing graphs. By applying diverse routing algorithms (e.g. Dijkstra, A* etc.) on these graphs, optimal routes (according to distinct requirements) can be calculated. That is, the benefit of this dissertation project is on the one hand the extension of OSM for indoor environments as well as the automated generation of standardized 3D building models from OSM. On the other hand, another benefit is also the generation of routing graphs from standardized city models. First research on indoor routing graphs has already been developed and published (Cf. [2]). Additionally, initial thoughts for generating city models based on OSM are available (Cf. [3]), as well as extending OSM to indoor environments (Cf. [4]).

[1] Goetz M., Zipf A. (2010): Open Issues in Bringing 3D to Location Based Services (LBS) ‐ A Review Focusing on 3D Data Streaming and 3D Indoor Navigation. 5th 3D GeoInfo Conference. Berlin, Germany.
[2] Goetz, M., Zipf, A. (2011): Formal Definition of an User‐adaptive and Length‐optimal Routing Graph for Complex Indoor Environments. In: Geo‐spatial Information Science (GSIS), Vol. 14, Issue 2. Springer.
[3] Goetz, M., Zipf, A. (2011): Towards Defining a Framework for the Automatic Derivation of 3D CityGML Models from Volunteered Geographic Information. Joint ISPRS Workshop on 3D City Modelling & Applications and the 6th 3D GeoInfo Conference. Wuhan, China
[4] Goetz M., Zipf A. (2011): Extending OpenStreetMap to Indoor Environments: Bringing Volunteered Geographic Information to the Next Level. 28th Urban Data Management Symposium. Delft, the Netherlands