Systematic Error Reduction in UAV-Based Laser Scanning by Using RTS

Abstract. In the last decades, Uncrewed Aerial Vehicles (UAV) with Light Detection and Ranging (LiDAR) sensors have become very popular for capturing high-resolution 3D-Point clouds, enabling the efficient measurement of large-scale objects in a short time. This is realized by fusing data from multiple sensors, usually Inertial Measurement Unit (IMU) and Global Navigation Satellite System (GNSS) to obtain the trajectory of the UAV and the LiDAR Scanner for point cloud generation. 
A crucial part in this process is the absolute positioning with GNSS. Systematic errors can occur especially due to challenging GNSS conditions, regarding the number of satellites, their distribution and site-dependent effects. These errors have a direct influence on the point cloud quality. 
The contribution of this paper is an extension of a UAV laser scanning system with a prism, that is continuously tracked using a Robotic Total Station (RTS). A factor graph-based trajectory estimation technique is used to fuse IMU and RTS data for a highprecise trajectory estimation to reduce systematic errors. The acquired data and a reference data set are used to evaluate our approach. The results show that point cloud misalignments can be reduced by integrating RTS data in the UAV trajectory estimation by up to 6 cm.