Unmanned Systems Technology 018 | CES show report | ASV Global C-Cat 3 USV | Test centres | UUVs insight | Limbach L 275 EF | Lidar systems | Heliceo DroneBox | Composites

76 Focus | Lidar Data processing The sensors are only part of the challenge though. In particular, for airborne applications such as mapping, the sensors are integrated into monitoring and measurement systems with software for analysis. This integration takes in a range of sensors according to the application, from image resolution to the flying height or range, coupled with data from GNSS satellite navigation and inertial measurement units (IMUs) to provide accurate positioning data. The expertise of the company doing the integration is in providing an overall system calibrated on the UAV using the IMU and GPS data from the scanner. That allows each point in the point cloud generated by the Lidar sensor to be referenced with accuracy and precision. Cloud analysis Operators are now adding cloud-based systems that allow fully automated data processing. Historically, the large quantities of raw data generated by laser-based mapping systems has made processing resource-intensive and costly. Now though, the raw data is uploaded directly to a cloud service in industry-standard Lidar formats and downloaded as processed data files in standard LAS format. This avoids the manual and highly technical post-processing procedure by using three automated steps, or ‘pipelines’. The first, the NavLab pipeline, is a trajectory module that processes inertial and GNSS data in forward and reverse several times to achieve maximum accuracy. The trajectory is then fused with 3D x, y and z Lidar data to create an unfiltered LAS file, which can be further refined using other processing pipelines, such as flight line matching and filtering. The final product is a calibrated and ‘de-noised’ point cloud with basic ‘ground/non-ground’ classification. The data can then be downloaded for further processing or forwarded to customers as a final deliverable. The modules can be used separately. NavLab for example can be used as a standalone module purely for trajectory processing, or as the starting point for additional pipelines such as planimetrics (2D mapping), advanced classification of objects or a digital terrain model. Another popular technology for surveying and mapping is photogrammetry, which uses camera images to determine the distance between objects. When used from a UAV, Lidar addresses some of photogrammetry’s challenges such as a lack of precision, an inability to penetrate vegetation and limitations in poor light. February/March 2018 | Unmanned Systems Technology Lidar sensor engines are integrated into systems for mapping from UAVs (Courtesy of Yellowscan)