UAV LiDAR–based 3D point cloud data processing for monitoring consolidation settlement during construction: a case study in Busan, South Korea

Abstract Construction sites with deep soft deposits experience a considerable amount of consolidation settlement. Settlement monitoring is important to prevent post construction settlement, which can lead to the tilting and overturning of structures. Generally, settlement plates are strategically positioned on site. However, their numbers are limited because of financial constraints. This limitation can result in unexpected settlement in unmonitored areas. Therefore, this study proposes a state-of-the-art approach for the monitoring of consolidation settlement utilizing unmanned aerial vehicle light detection and ranging (UAV LiDAR) technology. Busan Newport was selected as the study site, and measured 3D point cloud data were preprocessed to create a digital elevation model (DEM). Statistical outlier removal was selected as the optimal method for denoising data by comparing three methods on three ground types. The optimal grid size was then established to be 50 cm × 50 cm by evaluating UAV LiDAR data against GPS and total station measurements. Cloth simulation filtering, with parameters selected on the basis of a Cohen’s kappa coefficient of 0.9, was employed for bare-earth filtering. The distribution of consolidation settlement was visualized, capturing ground elevation changes via the DEMs of differences. The distribution of monthly settlement rates after the completion of the final loading revealed a deviation of 0.25 m. This finding indicates the potential for differential settlement and highlights the advantages of UAV LiDAR over practical monitoring techniques. The results of this work provide insights on settlement monitoring at construction sites with deep soft deposits.