Civil Engineering Departmental Seminar - Generation of initial digital twins using artificial intelligence
Generation of initial digital twins using artificial intelligence
Digital twins are increasingly used for the efficient operation of buildings and infrastructures. For a large number of existing buildings, digital models that can be used as a basis for digital twins do not yet exist. For the initial creation of digital twins, heterogeneous analog (e.g. drawings, documents, etc.) and digital (e.g. images, point clouds, databases) sources have to be systematically evaluated and transformed into digital models. In the context of the presentation, different artificial intelligence methods will be introduced that can be used for the generation of initial digital twins. This includes methods for the evaluation of conventional 2d drawings, point clouds, images and text information. The individual AI services can be executed in the form of service chains to generate geometric information in the form of 3D models and subsequently enrich these results with semantic information. Practical examples will be used to explain the results of an AI based generation in detail and challenges of application will be discussed. The research is funded by the German Ministry of Economics as part of the German AI Innovation Initiative.
Markus König is Professor for Computing in Engineering at Ruhr University Bochum, Germany since 2009. He studied civil engineering with a specialization in applied computer science at the University of Hanover, where he also received his PhD. Afterwards, he became an Assistant Professor for Theoretical Methods of Project Management at the Bauhaus University Weimar. He is currently Deputy Head of the National Center for the Digitalization of the German construction industry (BIM Germany). For his scientific and practical achievements, he was awarded with the Konrad Zuse Medal of the German Construction Industry Association in 2020.