ABSTRACT
Accurate detection and characterization of tension cracks for stability analyses purposes in open-pit mines requires high-resolution terrain models of the mine. This paper presents several contributions to the field of remote sensing of mines using Unmanned Aerial Vehicles (UAVs). A multivariate linear regression model for battery power consumption of drone has been derived by conducting an empirical study in various flight scenarios. The model has also been validated using data from a test flight. A genetic algorithm for solving the problem of flight planning and optimization have been proposed. The developed power consumption model has been used as the fitness function in this algorithm. A novel Android application has been developed for autonomous drone flights to follow mine terrain and capture high-resolution images. These images will be stitched together to be used for crack detection. A case study has been presented showcasing the ability of the developed software to achieve high-resolution imaging of an area.
