New possibilities for damage prediction from tunnel subsidence using aerial LiDAR data

Computation modelling has not been fully exploited for predicting building damage due to tunnel-induced subsidence, because of the expense and time required to create computational meshes for the vast quantity of buildings that may be impacted along a tunnel’s route. A possible circumvention of such a resource commitment lies in the exploitation of remote sensing data in the form of aerial laser scans (also know as Light Detection and Ranging – LiDAR). This paper presents work accomplished to date in the creation of a pipeline to automate the conversion of aerial LiDAR point cloud data directly into Finite Element Method (FEM) meshes without the intermediary step of triangulation-based conversion or reliance on geometric primitives through a Computer Aided Design (CAD) program. The paper highlights recent advances in flight path planning, data processing, plane identification, wall segmentation, and data transformation.