Using Multiple Open-source Low-cost Unmanned Aerial Vehicles (uav) for 3d Photogrammetry and Distributed Wind Measurement

Small, low-cost unmanned aerial vehicles (UAV) has made data acquisition more convenient and accessible for many applications. Using multiple UAVs (a coven) brings even more advantages like redundancy and distributed information. The objective of this paper is to show how a coven of UAVs can help two applications: measuring wind and 3D photogrammetry. INTRODUCTION Small, low-cost unmanned aerial vehicles (UAVs) are increasingly being used in various projects and applications to reduce cost, increase availability and to get real-time quality data. For example, Graml and Wigley [1] proposed a method to detect hotspots around the perimeter of a recently extinguished bushfire using a small IntelliTech Microsystems Vector-P UAV. By detecting these hotspots, this method can help firefighters to prevent the bushfire from reigniting. Another paper explains how EisenBeiss et. al. [2] used a helicopter with a CCD camera along with data from a laser scanner to create a 3D model of an archaeological site in Peru. The remote location and rugged topography usually makes getting to and working at the site difficult, however the small UAV proved to be a good tool to offset these drawbacks. Small UAVs are also very popular in a military setting for target detection and localization [3–5]. Barber et. al. [3] develops a method to detect and localize a target by circling around it, reiteratively finding its location, and using this data to find a better estimate. This method proved to be successful by decreasing the localization error from 40m to 5m. In ecological applications, we have shown how small UAVs can be used for soil moisture and habitat mapping [6]. All the advantages of small, low-cost UAVs are increased when using multiple UAVs (a coven) to achieve one goal. These advantages are shown by Drake et. al. [7] who used a coven to detect the location of a radar unit for military use. They used virtual vector fields to hold good geometry between aircraft, avoid collision, avoid no fly zones and circle the radar while localizing it. The location of the radar is calculated by comparing the time difference the radar signal reaches the different members of the coven. This triangulation method is called time difference of arrival (TDOA). Because the TDOA must be measured concurrently from different locations, successful and accurate radar detection may not be possible without more than one UAV. In another target detection and localization application, Pachter et. al. [4] shows how better localization accuracy can be achieved when using a coven of UAVs. While flying two different planes independently, each plane measured the location of the target with an error of about 10m. While working together, they were able to decrease the localization error to 5m. The objective of this paper is to exploit the advantages of small, low-cost UAVs for two applications: measuring wind and 1 Copyright c © 2009 by ASME Proceedings of the ASME 2009 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference IDETC/CIE 2009 August 30 September 2, 2009, Sa Di go, California, USA