The final meeting of the research project took place on December 17th. LiDAR4HVAC The presentation and discussion of the final project results will take place. Together with Research Burgenland GmbH, rmDATA GmbH, Spitzer GesmbH as well as Reiterer & Scherling GmbH The research project's objectives were successfully achieved.

As part of the collaboration, a digital workflow for the 3D reconstruction of building services components (BSE) based on LiDAR technologies was developed. This approach enables the geometric and semantic modeling of building services components (e.g., radiators, pipes) for renovation, planning, and execution in construction projects. The focus was on (semi-)automatic object recognition using machine learning and deep learning in combination with laser scanning for the efficient localization, registration, and segmentation of geometric data. Over the past three years, this has been achieved. AI models trained for object recognition and object segmentation for individual building services engineering (TGA) components and in corresponding Use cases tested.

Throughout the entire project duration, the rapid development in the field of Artificial Intelligence continuously monitored and specifically integrated into the research project. The results obtained represent a significant contribution to the inventory analysis and show that in the future both a higher planning accuracy as well as a significant reduction in time expenditure are possible for inventory analyses.

Furthermore, the developed software also addresses the following: Quality assurance and the associated Inventory surveys and analyses new standards.

Photos © Woschitz group