UAV Drone Lidar—Improving Point Cloud Data Quality via Smoothing

Martin Flood

Drone lidar—the use of compact, lightweight lidars on small unmanned systems—is an active area of research and commercial development. The use of drone lidar has accelerated over the past five years with the opening-up of the air space to commercial drone work. These changes in the regulatory environment, along with advances in lower-cost lidars, primarily developed for the automotive industry, the availability of multi-channel lidars (8-16-32-64-128 channels) and the development of alternatives to mechanical, rotating or oscillating mirror systems, have made drone-based lidar mapping affordable and efficient for many projects that would not be cost-effective with traditional fixed-wing or helicopter surveys. Drone surveys also offer advantages over static scanning in terms of coverage and remote accessibility for many project sites. This latest innovation in lidar is a continuation of the ongoing shift in the commercial lidar industry from “big, heavy, expensive” to “small, light, cheap”.

At GeoCue Group we design integrated lidar/camera payloads for use on small commercial drone platforms such as the DJI M600 or the Harris Aerial H6. Our focus is on high-accuracy data collection and processing for base mapping using tightly integrated lidars and mapping cameras in a single sensor. We refer to this as a 3D Imaging Sensor (3DIS) for its combination of high-accuracy elevation data with high-resolution oblique imagery, facilitating the simultaneous generation of a true 3D colorized point cloud and corresponding orthophoto. We focus our research and field testing on how to achieve the highest-accuracy results using relatively low-cost technology, with the goal of making this technology as widely accessible as possible to the survey and mapping community. Under our True View brand, we provide the results of our research and field work as end-to-end hardware and software solutions for researchers and commercial survey and mapping firms.