Energy efficiency, understood as obtaining the same output while consuming less energy, has become a central concern in recent years. Within this effort, the building sector deserves particular attention, as it alone accounts for approximately 30% of global final energy consumption, largely driven by heating and cooling demand. Any meaningful retrofit strategy, however, depends on reliable measurements capable of identifying where the intervention should be concentrated. This thesis proposes a multi-sensor framework that, through sensor fusion, combines the contributions of a LiDAR, a radiometric thermal camera, and a stereo RGB camera mounted on an UGV (Unmanned Ground Vehicle). The system enables the semi-automatic reconstruction of a geometric model of the building interior and the automatic, spatially resolved computation of the thermal transmittance (U-value) across different regions of a single surface, rather than assuming a homogeneous value per wall. The framework further defines a workflow for translating the reconstructed environment into an EnergyPlus model, so that the surveyed building can be simulated with as-built, spatially varying U-values, instead of the pre-defined constructions available in the software library. The resulting model provides a more faithful representation of the actual thermal behaviour of the envelope, supporting subsequent energy-demand analysis.
L'efficienza energetica, intesa come la capacità di ottenere lo stesso risultato consumando meno energia, è diventata negli ultimi anni una questione centrale. In questo contesto, il settore edilizio merita particolare attenzione, poiché da solo rappresenta circa il 30% del consumo finale globale di energia, in gran parte dovuto alla domanda di riscaldamento e raffrescamento. Qualsiasi strategia di riqualificazione efficace, tuttavia, dipende da misurazioni affidabili in grado di individuare dove concentrare l'intervento. Questa tesi propone un framework multi-sensore che, attraverso la sensor fusion, combina i contributi di un LiDAR, di una termocamera radiometrica e di una camera RGB stereo montati su un UGV (Unmanned Ground Vehicle, veicolo terrestre senza equipaggio). Il sistema consente la ricostruzione semi-automatica di un modello geometrico degli interni dell'edificio e il calcolo automatico, risolto spazialmente, della trasmittanza termica (U-value) su diverse regioni di una stessa superficie, anziché assumere un valore omogeneo per ciascuna parete. Il framework definisce inoltre un flusso di lavoro per tradurre l'ambiente ricostruito in un modello EnergyPlus, in modo che l'edificio rilevato possa essere simulato con U-value as-built e variabili nello spazio, invece delle stratigrafie predefinite disponibili nella libreria del software. Il modello risultante offre una rappresentazione più fedele del comportamento termico effettivo dell'involucro, a supporto della successiva analisi del fabbisogno energetico.
Robotic thermal mapping of building facades: a multi-sensor rover platform for automated environment modelling and thermal bridge detection
WIDMANN, LORENZO
2025/2026
Abstract
Energy efficiency, understood as obtaining the same output while consuming less energy, has become a central concern in recent years. Within this effort, the building sector deserves particular attention, as it alone accounts for approximately 30% of global final energy consumption, largely driven by heating and cooling demand. Any meaningful retrofit strategy, however, depends on reliable measurements capable of identifying where the intervention should be concentrated. This thesis proposes a multi-sensor framework that, through sensor fusion, combines the contributions of a LiDAR, a radiometric thermal camera, and a stereo RGB camera mounted on an UGV (Unmanned Ground Vehicle). The system enables the semi-automatic reconstruction of a geometric model of the building interior and the automatic, spatially resolved computation of the thermal transmittance (U-value) across different regions of a single surface, rather than assuming a homogeneous value per wall. The framework further defines a workflow for translating the reconstructed environment into an EnergyPlus model, so that the surveyed building can be simulated with as-built, spatially varying U-values, instead of the pre-defined constructions available in the software library. The resulting model provides a more faithful representation of the actual thermal behaviour of the envelope, supporting subsequent energy-demand analysis.| File | Dimensione | Formato | |
|---|---|---|---|
|
WidmannLorenzo_PDFA_v2_compressed.pdf
embargo fino al 05/10/2029
Dimensione
20.4 MB
Formato
Adobe PDF
|
20.4 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/116257