The International Association for Automation and Robotics in Construction
Publications / CCC 2025 - Zadar, Croatia
MASSIVE DATA CAPTURE MODELING FOR THE REFURBISHMENT OF EXISTING BUILDING STOCKS
The rehabilitation of buildings is a pressing global requirement that could contribute to reduce the negative economic, environmental and social impacts of the construction sector. Current documentation on existing obsolete building stock is missing or outdated. Therefore, digital models need to be generated, such as creating building information models from scratch. However, this has limitations, for example that the resulting model is an abstract theoretical version that cannot reproduce the geometric singularities of the real building, which has implications for structural performance among others. Thus, the automation of this digitalization is an interesting solution that has been investigated for years. This research paper presents a novel approach for the digitalization of non-heritage buildings that reduces data capture timings but is precise enough for retrofitting applications. The new approach uses laser scanner, thermal infrared sensing, high quality pictures (HQP) and automatic frame extraction (AFE) from video. Data preparation for the 3D reconstruction is the main novelty, which has been applied to obtain the surroundings and building information model (BIM) of the reference building for Barcelona schools. Findings coincide with previous projects regarding the high accuracy of the laser scanner and the broad coverage of photogrammetry. New results qualify HQP as a highly efficient method. Its combination with AFE increase the coverage to high levels. The proposed approach is part of the project "Waste-based Intelligent Solar-control-devices for Envelope Refurbishment" starts, of "Ecological Transition and Digital Transition Projects" of the Spanish Ministry of Science and Innovation (MICINN) with reference TED2021-130155B-I00, funded by MCIN/AEI/10.13039/501100011033 and by the European Union "NextGenerationEU"/PRTR. This proposal surpasses the manually modeled BIM errors and enables digitalizing architectural clusters to generate urban digital twins to enhance the proper future management of urban stocks.