LiDAR technology for the reconstruction of technical building installations and systems

The refurbishment of existing buildings is of crucial importance for achieving climate targets. However, in order to accelerate the rate of refurbishment, time and cost-efficient methods are required to record the existing heating and air conditioning systems.

LiDAR technology (Light Detection and Ranging) is seen as having great potential for the future to enable automated recording of existing geometries. Automated segmentation of the point cloud in relation to technical building installations and systems is not yet possible. Accordingly, LiDAR applications are still associated with a high level of manual effort. The provision of relevant information at the right time on the basis of countless data sources plays an important role here. In this project, a method for the 3D reconstruction of technical building components for geometric and semantic modelling in the refurbishment, planning and execution phase of technical building equipment based on LiDAR technology is being developed. The focus is on object recognition using (semi-)automatic approaches such as machine learning, deep learning in combination with laser scanning and photogrammetry methods for efficient localisation, registration and segmentation of geometric data and objects. Furthermore, minimum requirements for as-is conditions and process models are developed that are used in digital working methods (BIM) for the exchange and recording of information in the planning and execution of refurbishment projects. The methods and procedures developed will be tested by the company partners on the basis of exemplary use cases.