Uncrewed Aerial Systems (UAS) are a tool used to help capture remote sensing data of an area. Commonly referred to by the general public as drones, UAS is a great and accessible way for an aerial machine to get imaging or data points of a small area of the Earth’s surface without having to touch down, making it very useful for difficult to traverse or dangerous areas.
How UAS works
UAS refers to an aircraft where there is no human pilot. The craft is either autonomous with a pre-programmed GPS route or is remotely operated. The two main pieces are the aerial vehicle and the ground-based controller. The team on the ground has the communication and controls, as well as sometimes processing and analysis tools.
To track where it is, they use the Global Positioning System (GPS), giving the researchers an accurate and precise location at all times. The ground station can communicate with the UAS through satellite communication or radio waves to gather data and send commands and directions. Some more advanced UAS have sensors to detect and steer away from obstacles in their paths, recalibrating as they encounter them or having the remote controller avoid any hazards.
Why use UAS?
There are many models of UAS available, each with their own pros and cons. Smaller ones are useful for creating 3D models. Larger UASs are more expensive but have the ability to carry additional instruments like lidar, SAR, magnetometers, and thermal or spectral imaging cameras.
Being both uncrewed and aerial allows for UAS to fly over places where it is too dangerous to approach on the ground. Inaccessible areas can still be observed and analyzed with the use of UAS. Additionally, compared to monitoring methods on the ground, using a UAS can be a less invasive and more environmentally-friendly way to observe an area without disturbing or disrupting the natural surroundings.
Structure from Motion
One technique that UAS is used for is called Structure from Motion (SfM). This is an imaging technique that combines photos from a moving camera to create 3D models of areas. The UAS takes multiple 2D photos of the area, and then by overlaying multiple scans from multiple angles, a highly accurate topographical model can be created.
This differs from the way that something like a Terrestrial Laser Scanner generates a 3D model. Rather than measuring distances with a laser, Structure from Motion relies on parallax. When comparing photos taken in different locations, points will move more or less depending on their distance from the camera.
The only required tools are a camera and some processing software, making SfM a relatively accessible technique. (Although the addition of position markers precisely measured by GPS instruments enables much more accurate models.) While it is useful for mapping topography, SfM does not require a UAS — it can also be collected from a handheld camera or an airborne platform other than a UAS like a tethered balloon or kite. This allows a range of size of 3D models, from decimeters to several kilometers, with the limitation being the larger the models and the more images, the longer the processing time for model generation.
Applications
Since many different kinds of cameras and sensors can be attached and the features are always evolving and advancing, the applications of UAS are large and vast. Over long periods of time, UAS can help to model the evolution of landscapes and how they are moving. One study did this in 2021 for the Selmun promontory in northern Malta. They were able to characterize the area by gathering data about the spatial geometry of the landslide to support monitoring and early warning systems for the area.
Additionally, combining UAS and SfM can generate digital elevation models to have a variety of applications like quantifying erosions, landslide deformations, earthquake faulting, and glacial retreats. SfM can assist UAS in tracking evolutions of areas over both short and long periods of time. One group used a combination of the two to survey and map the Home Hill Landslide in Tasmania over a larger time period, while also using it to track active lava flow from the Sinabung Volcano in Sumatra, Indonesia.
Lots of data is open access and available to researchers, and many can use that as comparison points if they personally were not monitoring the area years ago. One study used multiple previous study’s data throughout 10 years prior to produce a data set to measure and track strain across the San Andreas fault at Dry Lake Valley in California.
UAS and SfM are versatile, and can cover large areas that without an aerial craft would be extremely time-consuming. One study took advantage of these tools’ unique abilities and monitored volume changes of the Exit Glacier in Alaska, covering 100–300 square km in a day.
A catalog of OpenTopography datasets using SfM and UAS can be found here.