A new approach for monitoring mining surface 3D deformation using UAV-LiDAR point cloud data

Abstract

This study introduces a method that leverages Unmanned Aerial Vehicle – Light Detection and Ranging (UAV-LiDAR) point cloud to monitor the three-dimensional surface deformation of a mining face accurately. The method incorporates Progressive TIN Densification (PTD) filtering to distinguish between ground and non-ground points within the UAV-LiDAR dataset. It then uses various techniques to pinpoint feature points, which are aligned using Iterative Closest Point (ICP) registration algorithm to determine corresponding points and calculate the resulting deformation of the surface caused by mining activities. By applying this deformation extraction method, two sets of UAV-LiDAR point cloud data collected at different times over the Wangjiata Coal Mine in Inner Mongolia were analyzed to evaluate three-dimensional surface deformation. Results indicate that the surface movement in the x-direction ranged from −0.69 m to 0.64 m, in the y-direction from −0.58 m to 0.53 m, and in the z-direction from −2.79 m to 0.2 m. Additionally, the refined three-dimensional deformation model demonstrated interpolation accuracies of 40 mm, 43 mm, and 74 mm in the x, y, and z directions, respectively and the external coincidence accuracy is 51 mm, 48 mm, and 55 mm respectively. This method effectively harnesses the potential of the collected point cloud data, offering a viable solution for extracting three-dimensional deformations of mining faces through UAV-LiDAR monitoring, with promising application prospects.