Optimal placement of precast bridge deck slabs with respect to precast girders using 3D laser scanning

Abstract Because precast components offer faster production, lower cost, and more efficient construction, more and more bridges are constructed using precast components rather than conventional on-site construction. For example, precast bridge deck slabs are placed on precast girders, and they are connected by shear pockets on deck slabs and shear connectors on girders. For the coupling of deck slabs and girders, it is important to ensure proper connections between shear pockets and shear connectors. However, shear pockets and shear connectors often do not match properly at construction sites because of dimensional errors and misalignments. Furthermore, precast girders deform over time due to their heavy weights, time dependent creep and shrinkage, pre- or post-tensioning, etc., once they are placed on sites. To match these components at construction sites, workers often need to trim and cut some components, delaying construction processes. To shorten such delay, this study proposes a laser scanning and signal processing technique that can automatically identify the optimal placement of precast bridge deck slabs with respect to precast girders by minimizing mismatches between shear pockets and shear connectors. First, scan data from precast bridge deck slabs and precast girders are acquired using a 3D laser scanner, and their dimensions are estimated including the locations and dimensions of shear pockets using DBSCAN and mixed pixel filtering algorithms. Next, locations of the shear connectors are extracted from the scan data of the precast girders using RANSAC and K-means clustering algorithms. Finally, the optimal placement of the deck slabs with respect to the girders is determined by solving a nonlinear minimization problem considering the locations and sizes of the extracted precast components. To validate the performance of the proposed technique, experiments were conducted on small-scale test specimens and at an actual construction field. The experimental results demonstrated that the proposed technique could effectively estimate the optimal placement of precast bridge deck slabs with respect to precast girders.