Development of a UAV-based system for the semi-automatic estimation of the volume of earthworks

Abstract

Abstract One of the challenges faced by surveyors in acquisition of accurate spatial data for mining applications is the risk involved in acquiring data in rugged terrains and difficult or inaccessible areas. With the advent of modern technology, accurate geospatial data can now be safely obtained for proper mining documentation periodically. The use of Unmanned Aerial Vehicles (UAVs) for data acquisition in mine surveying has been a viable means of obtaining reliable geospatial data rapidly and efficiently. The main goal of this study is to develop a semi-automatic UAV-based system for the acquisition of spatial data required for the estimation of the volume of earthworks. A DJI Phantom 4 quadcopter was used for the acquisition of image data of the project site, while the images were processed into a Digital Elevation Model (DEM) using Pix4Dmapper v2.0.1, which was then imported into the MATLAB-based system developed for the automatic estimation of the volume of earthworks. The volume obtained from the automated system was thus compared with the volume obtained directly from the Pix4Dmapper software, having specified a contour interval of 1 and an allowable error rate of ±3% as the standard error. While ±1.02% error was observed in the volume estimated using the Pix4Dmapper, the developed automated system yielded an estimated precision of ±0.81% in its volume estimation, which proves to be more robust for automatic volume estimation in terms of accuracy and precision.