Banner Img

© Thomas Stastny, All rights reserved

Banner Img

© Thomas Stastny, All rights reserved

Thumbnail Image
Thumbnail Image

Autonomous Deployment of GNSS Stations on Polar Outlet Glaciers Using a Long-Range, Tilt-Wing UAV

Thomas Stastny

-

ETH Zurich

Lay summary

Studying the dynamics of fast-flowing outlet glaciers and ice streams in polar regions requires the acquisition of data from some of the most inhospitable, dangerous, and remote environments in the world; a challenging, but necessary endeavor to develop physical models for reliable predictions of the stability of the polar ice sheets and the resulting global sea level rise. In situ measurements are mandatory for ground truth validation of models and observation of higher temporal resolution events (hour/minute scale) on fast-moving glaciers. However, present day installation of measurement devices on large-scale polar glaciers is a tedious, dangerous, and costly task, often requiring helicopter transport to reach points that are otherwise inaccessible. Recent developments in aerial robotics have demonstrated the capability of multi-copter Unmanned Aerial Vehicles (UAVs) to automatically deploy and recover sensory payloads. However, these platforms are severely limited by flight time and range, and in general not applicable for the scale of large polar outlet glaciers. This project will tackle this limitation by combining the advantages of hover-capable multi-copter UAVs with the range and efficiency of fixed-wing UAVs, developing a novel, hybrid, tilt-wing UAV, capable of autonomous placement and recovery of low-power GNSS stations at rough, inaccessible positions on polarscale glaciers, tens of kilometers from base camp. At such long ranges from base, and in remote polar regions without wireless broadband data links, centimeter-accurate global positioning via Real-Time-Kinematic (RTK) corrections cannot be relied upon. To ensure precision payload drops, we will develop lightweight computer vision and machine learning -based approaches for the UAV to localize near the drop-site as well as with respect to the payload itself during efficient, forward-flight, high-speed payload recoveries. The system will be field tested at the Rhone Glacier, Switzerland, in Summer 2021, where varying degrees of glacier crevassing and elevation roughness will provide a range of test conditions. If successful, this technology will enable an immense capability for polar scientists to safely install and recover measurement devices at remote, inaccessible locations in polar regions, and at a fraction of the cost and carbon footprint of manned helicopter transports.

Details

Regional focus Arctic
Location Rhone Glacier, Switzerland
Funded amount 50,000 CHF
Project dates 15th February 2021 – 15th November 2022
Category SPI Technogrants
Keywords
glacier motion/ice sheet motion, remote sensing, long-range transport, UAV, in situ measurements