Investigating the applicability of unmanned aerial vehicles (UAV) photogrammetry for the estimation of the volume of stockpiles.

Abstract

Accurately estimating the volume of earthworks is very important in mining engineering and construction. This estimation can be difficult because of the morphological condition of the stockpiles, hence, devising simpler, yet accurate methods of stockpile volume estimation is still a research problem in mining. Two non-invasive survey methods were compared in this research: the conventional ground-based and UAV-approach, for the survey of a twin-stockpile of gravel using Leica TS06 Total Station and DJI Mavic Air UAV, respectively. About 128 images of the area were acquired at 50 m flying height and 75% overlap during the flight mission. The images were processed using Agisoft Metashape Pro; a digital photogrammetric software, and the DEM obtained was used for the volume estimation. The total station data was also processed in ArcGIS to generate a TIN-model from which the volume was also estimated. The volume estimated from the TIN-model was compared with the volume estimated from the UAV-based DEM, using the volume obtained from the mill-machine as the standard. The obtained result shows that while 2750 m³ was obtained as the cumulative volume from the mill machine, the UAV approach yielded 2686.252 m³ and the ground survey approach gave 2830.713 m³. The percentage difference between the two methods compared to the actual volume is 2.94% and −2.31%, respectively. These results, and the result of the processing time analysis show that UAV approach is both accurate and time economical, which attests to the potentials of low-cost UAVs to provide robust alternative to the time-consuming and rigorous ground survey approach.