Project description

In order to increase the usability of CityGML for more user groups and areas of application, the OGC CityGML Standards Working Group (SWG) and the Special Interest Group 3D (SIG 3D) of the initiative Geodata Infrastructure Germany (GDI-DE) work since 2014 on the further development of CityGML.

This development results in the new major version CityGML 3.0 which is to be finalised by the end of 2022. The new version brings a number of improvements, extensions, and new functionalities.

• Use of a model-driven approach: The data model is now fully based on the ISO 191xx standards.The software tool Enterprise Architect is used for creating the data model and the GML encoding will be fully automatically derived from the data model using the software tool ShapeChange.
• Revised LOD concept: LOD4 has been removed, the interior of objects can be expressed in different LODs 0-3 now. It is e.g. possible to model the outside shell of a building in LOD1 while representing the interior structure in LOD2 or 3.
• New and revised modules: CitygML 3.0 includes several new and revised modules which allow for better use of 3D city models in different areas of application such as urban planning, energy simulations, and traffic analyses, reduce redundancy by integrating and providing concepts used in several modules centrally via inheritance, simplify geometry handling, and improve the representation of physical and logical objects.
  • New Core model: All spatial representations base on the two new pivotal concepts Space and SpaceBoundary. The geometry can now also be given by point clouds. The new core model implements the new LOD concept.
  • New Construction module: Groups all classes which are similar over different types of constructions like buildings, tunnels, and bridges. Allows for representing constructions not being covered by any of the other modules.
  • New Versioning module: Allows to explicitly represent and exchange multiple versions of city objects (e.g. history or alternative designs) within one city model dataset. All objects can have bitemporal lifespan data.
  • New Dynamizer module: Defines concepts to represent and exchange time-varying data for city object properties (e.g. solar irradiation over the day) as well as to integrate sensors with 3D city models. Data sources can be timeseries data represented inline, in external files, or from sensor web services (OGC Sensor Web Enablement or Internet of Things, IoT, in general).
  • Revised Transportation module: Transportation objects like Roads, Tracks, or Railways can now be subdivided into sections and have an area as well as a center line representation for each LOD.
  • Revised Building module: Allows now for representing storeys and building units.
• Improved interoperability with other standards: CityGML 3.0 allows for better interoperability with other relevant standards like IndoorGML, Industry Foundation Classes (IFC), Land Administration Domain Model (LADM), as well as with Semantic Web Technologies like Resource Description Framework (RDF) and with the EU Directive INSPIRE (Infrastructure for spatial information in Europe).
• Backwards compatibility: All modifications to the new CityGML 3.0 model are carried out in a way to ensure backwards compatibility with CityGML 1.0 and 2.0

The CityGML 3.0 standard consists of at least two parts:

• The specification "CityGML 3.0 Part 1: Conceptual Model Standard"
• The specification "CityGML 3.0 Part 2: GML Encoding Standard"
• Further encoding specifications (e.g. relational database schema, JSON-based representation) which may follow in the future

Part 1 of the standard has been published on 13 September 2021 by the OGC: https://docs.ogc.org/is/20-010/20-010.html

Part 2 of the standard is currently being finalised by the OGC. Publication is planned for the end of 2022.

Please note that all changes are subject to the final voting of the OGC CityGML SWG and the OGC Technical Committee.

Conceptual Model

The conceptual model of CityGML 3.0 was defined as UML model. The UML model was created using the software Enterprise Architect. The UML model can be downloaded from this github repository: https://github.com/opengeospatial/CityGML-3.0CM/releases/tag/3.0.0-final.2021.02.23 (file "CityGML_3.0_Consolidated_Draft.eap")

GML Encoding

GML application schemata have been derived from the UML model using the software ShapeChange. The current draft of the GML application schemata can be downloaded from this github repository: https://github.com/opengeospatial/CityGML-3.0Encodings (Folder "CityGML")

Test data sets

Test data sets that are compliant to the CityGML 3.0 standard are available here:

• Test data sets for the individual CityGML modules: https://github.com/opengeospatial/CityGML-3.0Encodings/tree/master/CityGML/Examples
• Various test data sets that have been created by mapping publicly available CityGML 2.0 data sets to CityGML 3.0 (e.g. Rotterdam, Helsinki, KIT-Railway): https://1drv.ms/u/s!Ag_9VT-F89-7jlqoc77pjKmG-8VP?e=G551gC
• Test data sets for the Transportation module: https://github.com/tum-gis/citygml3.0-transportation-examples
• A test data set that makes use of the newly introduced class "BuildingConstructiveElement": https://github.com/tum-gis/ifc-to-citygml3 (Folder "output")

 

Software support

The following software tools already support CityGML 3.0:

• citygml2-to-citygml3 - A converter for mapping CityGML 2.0 files to CityGML 3.0: https://github.com/tum-gis/citygml2-to-citygml3
• ifc-to-citygml3 - An FME workspace for converting IFC files to CityGML 3.0: https://github.com/tum-gis/ifc-to-citygml3
• r:trån - A converter for mapping OpenDRIVE files to CityGML: https://www.asg.ed.tum.de/en/gis/software/rtron/
• citygml4j - An Open Source Java API for CityGML: https://github.com/citygml4j/citygml4j  
  (The support for CityGML 3.0 is currently in beta stage and available via this gishub branch: https://github.com/citygml4j/citygml4j/tree/citygml3-devel)
• FME by making use of the generic GML-Reader/Writer: https://www.safe.com/

Duration

2013-2022

Contact person

• Dr. Tatjana Kutzner

Project leader

• Prof. Thomas H. Kolbe

Project contributors

• Prof. Thomas H. Kolbe
• Dr. Tatjana Kutzner
• Kanishk Chaturvedi
• Dr. Andreas Donaubauer
• Christof Beil