ASSESSING THE ACCURACY OF UAV SURVEYS WITH DIFFERENT CONFIGURATIONS

Abstract. Recently, developments in airborne sensors and easy-to-fly, reliable, low-cost commercial UAVs have opened a new era for precise aerial mapping. The restricted payload capacity of low-cost UAVs imposes constraints on the quality of their navigation systems and the sensors they can carry. Therefore, the quality of photogrammetric products generated from UAV surveys needs to be assessed. In this study, photogrammetric mapping principles are employed in collecting, processing, and analyzing optical images collected using two UAV surveying systems. The first system was a 3DR Iris+ drone flying at 25 meters above the ground. The second system was a homemade drone similar to Tuffwing flying at about 70 meters above the ground. Both systems carried a Canon PowerShot S100 using a fixed 5.2 focal length. About 30 ground points were surveyed with a TS02 total station and served as the ground truth for data evaluation. After the flights, data was proceeded and geospatial products including DEMs and orthophotos produced were evaluated. Different ground control configurations were examined and ground check-points were used to evaluate the final accuracy of the geospatial products. The comparison of the derived 3D information from captured data with ground measurements showed a high correlation between the accuracy of the 3D products and the sensor specification, flying altitude, as well as image layout. This was supported by comparisons between actual errors and theoretical positional precision based on flying height, photo scale and air base was conducted.