[ad_1]
Planetary scientists at NASA’s Jet Propulsion Laboratory developed the machine learning tool to save researchers time by training AI with 6,830 images of the Red Planet.
It has already found a crater that formed between March 2010 and May 2012.
This crater was very small compared to others – only four meters in diameter – making detection difficult for human scientists.
It looks for events such as dust devils, avalanches and shifting dunes and has been used to find over 1,000 craters.
However, MRO images can usually only capture the signs of the explosion around the impact of a crater, rather than the crater itself.
Without artificial intelligence, scientists must then examine those images with the High-Resolution Imaging Science Experiment (HiRISE).
The activity takes time; it can take a researcher 40 minutes to properly scan an image.
This new AI, once trained, was set to analyze the entire library of 112,000 images taken by the Context Camera.
Running on a supercomputer, the AI is capable of detecting craters at a speed 480 times faster than humans, reducing the 40-minute detection time to just five seconds.
750 copies of the classifier were made concurrently. “It would not be possible to process over 112,000 images in a reasonable amount of time without spreading the work across many computers,” said Gary Doran, computer scientist at JPL.
“The strategy is to break the problem down into smaller parts that can be solved in parallel.”
Despite the achievements of artificial intelligence, it is still necessary for a human to check the accuracy of his work due to his inability to do more qualified analyzes.
“Tools like this new algorithm can be their assistants. This paves the way for an exciting symbiosis of human and artificial intelligence “researchers” working together to accelerate scientific discovery, “said Kiri Wagstaff, computer scientist at JPL.
Ultimately, the goal is for systems like these to run on computers aboard Mars orbits, rather than computers on Earth.
Currently, the data sent to Earth still requires scientists to examine it – a task of Michael Munja, a graduate student at Georgia Tech who worked on the classifier, compared to finding a needle in a haystack.
“The hope is that in the future, artificial intelligence can prioritize orbital images that scientists are most likely to be interested in,” Munje said.
It is also hoped that the instrument will offer a more comprehensive view of the frequency with which meteors hit Mars, as well as find even smaller impacts that have been overlooked by scientists.
“There are probably many other impacts that we haven’t found yet,” said scientist Ingrid Daubar.
“This advancement shows you how much you can do with veteran missions like MRO using modern analytical techniques.”
This is not the only case where artificial intelligence has been used by NASA to detect lost data from human scientists.
The 50 planets range from the size of Neptune to the smallest of Earth. Some had orbits lasting up to 200 days on Earth, while others revolve around their respective stars once a day.
Another AI tool was developed to reveal the structure of the universe by building a tool called the “Dark Emulator” that can create hundreds of virtual universes and use those simulations to help scientists find out more about our reality.
Source link
