Real-time and Autonomous Detection of Sea Ice and Growlers Using UAS
Abstract
Maritime activity in the high north is rapidly increasing. Even though the activities occur mainly during the summer months when sea-ice extent is minimum, icebergs and growlers from glaciers can still pose a danger. Growlers and small icebergs are often too small to be reliably detected by satellites, while at the same time they could be large enough to damage ships and infrastructure.
We have developed an unmanned aerial systems (UAS) for real-time detection and classification of drifting sea ice and growlers implemented on an electric fixed wing unmanned aircraft (UAV). The UAV has a wingspan of 3 meters, a takeoff weight of 10 kg including payload and batteries, and can operate up to 3 hours with a cruise speed of 22 m/s. The payload consist of a high resolution electro optical camera, an IR camera, a long range phased array broadband radio and a powerful single board computer. All recorded images are geocoded and analyzed in real time onboard the aircraft, where ice coverage and position, shape and size of all individual ice sheets are stored in a database onboard the aircraft. The database is then synchronized with all assets with access to the system, first with the ground station and then over local network or internet to other parties. We have developed a situational awareness tool called Nlive, which allows multiple users to simultaneously browse and interact with the results using a standard web browser. The system, which consists of both the UAS including the onboard processing and the Nlive dissemination platform has been demonstrated during two field campaigns in Kongsfjorden, Svalbard in the periods April 24-May 8, 2017 and April 12-26, 2018. During the campaigns we collected coordinated observations of growlers using different methods. All observations which contains geospatial and time-tagged information are included in the Nlive platform. During both campaigns the ground station was connected to internet and all the data from the UAV was synchronized with a server in Tromsø, Norway. Here, we present the Nlive system using data collected during the extensive field campaigns. We believe that efficient use of remote sensing methods, applying both terrestrial radar, satellite SAR, and unmanned aircraft systems (UAS), can provide and maintain situational awareness with regards to drifting sea ice and growlers.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.C31E1578S
- Keywords:
-
- 0750 Sea ice;
- CRYOSPHEREDE: 0758 Remote sensing;
- CRYOSPHEREDE: 1621 Cryospheric change;
- GLOBAL CHANGEDE: 4207 Arctic and Antarctic oceanography;
- OCEANOGRAPHY: GENERAL