Semi-automated Generation of Geometric Digital Twin for Bridge Based on Terrestrial Laser Scanning Data

Digital twins (DTs) have a great potential for bridge operation and maintenance. Geometric digital twins (gDTs) are the key component of DTs. At present, a growing number of researchers are using high-precision 3D laser point clouds to generate gDTs. However, for large bridges, such as arch, cable-stayed, and suspension bridges, comprehensive point-cloud collection stations are difficult to set up due to their large span, narrow site, and limited field of vision. Consequently, the complete point clouds of these bridges cannot be easily obtained. Thus, knowing how to process absence point clouds and generate gDTs is an urgent problem. This study proposes a semiautomatic method of extracting geometric information of a bridge's components in the absence of point clouds. First, an algorithm based on the combination of the iterative polynomial fitting curve and sliding window is developed to extract the arch ring accurately. Second, an improved random sample consensus (RANSAC) algorithm based on distribution density is adopted to extract the cross sections of the arch bridge components, except the arch ring. For cross sections that lack point clouds, a translation strategy is used to supplement the unknown line segment. Finally, for the T-beam, a model alignment method is proposed to best match the characteristic intersections extracted by the improved RANSAC algorithm and the points corresponding to the design model. The quality of the generated models is gauged using a point cloud deviation chromatogram. In addition, the stressed component piers are compared with its design parameters to verify the accuracy of the proposed method. Results show that our method can efficiently and accurately extract geometric information and generate gDT for the bridge.

Automated summary

Digital twins have great potential for bridge operation and maintenance, with geometric digital twins being the key component. However, the point-cloud collection for large bridges is challenging, leading to the problem of processing absence point clouds and generating accurate geometric digital twins.