Processes supported by laser scanning create important new sources of information for existing structures, because the original construction plans do not always exist or are inaccurate due to various renovations and complex interior fittings (especially in the case of pipe systems). This technology is also suitable for digitally gaining an overview of the current state of a construction site. Laser scanning devices (portable or static) are used to record spatial images of the structures as point clouds and refer to them locally.
On the basis of such point clouds, the chair of Computing in Engineering deals in current research topics with the automated generation of BIM-based models, the efficient processing of raw data and the comparison between real and virtual structures. Here, methods of Artificial Intelligence (AI) are in the foreground of the research. For example, patterns in the point clouds are determined and classified. Errors in scanning pose a challenge, such as shaky data sets or noise on reflective surfaces. At the chair of Computing in Engineering we try to develop algorithms with our research that allow the use of point clouds in BIM-based models.