This study integrates vision-based sensing and digital twin technologies into large-scale conformal 3D Concrete Printing (3DCP). By capturing and reconstructing non-planar, rapidly changing surfaces—emulating real-world construction conditions—the research enhances adaptability and real-time monitoring. Experiments demonstrate high-fidelity digital reconstructions (RMSE values of 1.76 mm on sand and 2.94 mm on gravel), enabling precise conformal printing with improved filament consistency. Additionally, the digital twin is used to detect and quantify discrepancies between planned and executed material deposition, strengthening process accuracy. This vision-augmented approach promises not only to refine 3DCP scalability and operational efficiency but also to bolster sustainability by reducing material waste in automated construction. Collaborators: Paniz Farrokhsiar, Sven G. Bilén, José Pinto Duarte, Benay Gürsoy
