Scientists use artificial intelligence to identify mystery sightings in the night sky
Scientists are using Artificial Intelligence to help them identify aerial phenomena in the night sky.
Using artificial intelligence, Technische Universität Ilmenau (Germany) is improving the detection and classification of unidentified phenomena in the night sky. In collaboration with the American Meteor Society, the group for data-intensive systems and visualization formed AllSky7, an international network of scientists and amateur astronomers that observes the night sky continuously with specially designed cameras and classes and assigns the events they observe.
Our night skies are often filled with luminous phenomena that make us wonder about the universe. There are some that we can explain: turbulence in the atmosphere creates sparkling stars, or glistening meteorites create shooting stars.
There are other surprises you may not expect: satellites passing overhead at breakneck speed or rocket engines falling back to Earth. A major goal of the AllSky7 network is to capture, detect, and classify all night sky phenomena around the world. Founded in 2018, the American Meteor Society encourages professional and amateur astronomers to collaborate on research through its international team.
In addition to tracking meteorites that are falling toward Earth, AllSky7 also aims to identify sky phenomena caused by other events. 360-degree special cameras continuously monitor the night sky across 85 sites in the US and Europe, detecting countless phenomena that are analyzed and classified by day by the camera operators. It should be noted that the algorithms were trained only for a few so-called positive classes, which meant they were insufficiently able to distinguish meteors from other kinds of events.
This algorithm and data were developed over a six-month period by Rabea Sennlaub and Martin Hofmann. With the help of the AllSky7 network, they collected the images of 20,000 meteors and non-meteors taken at Sonneberg in Thuringia, Germany, and further separated them into subclasses to ensure finer classification. Mike Hankey, a researcher from the U.S., says the Thuringia research is a giant step towards a seamless sky observation.
It is now possible to estimate with greater precision the amount of space debris that threatens communications satellites and the lives of astronauts aboard space stations. As a result of the findings, an international partnership is formed between wide-angle sky observatories around the world. Meteors can also be tracked by the network to determine where they are landing and when they fall to Earth. Using this method, we can analyze rock debris and learn more about the origins of our solar system.
Royal Astronomical Society’s Monthly Notices published the related research.