[ad_1]
It’s easy to find large meteorites (or their craters) once they reach Earth, but smaller ones are often overlooked – scientists get less than 2% of them. Soon, however, it might just be a matter of sending a robot to do the job. Universe today reports that the researchers have developed a system with autonomous drones using machine learning to find the smallest meteorites in impact sites that are either “hidden” (even though observers have traced the fall) or simply inaccessible.
The technology uses a mixture of convolutional neural networks to recognize meteorites based on training images, both from online images and staged shots from the team’s collection. This helps the AI to distinguish between space rocks and ordinary stones, even with a variety of shapes and terrain conditions.
The results are not flawless. While a test drone correctly spotted planted meteorites, there were also a few false positives. It could be some time before robotic planes are reliable enough to provide accurate results on their own.
The implications for space science are significant if the technology turns out to be correct, however. This would help scientists spot and potentially recover meteorites that are either too small to be found or too far away. This, in turn, could help locate sources of meteorites and identify rock compositions. Put simply, drones could fill the gaps in humanity’s understanding of the cosmic debris landing on our doorstep.
[ad_2]
Source link