mars-crater-artificial intelligence nasa ai

Image via Adobe Stock

Here’s how NASA’s new AI tool will help scientists discover craters on Mars

Scientists from NASA’s JPL will use AI and a machine-learning algorithm to speed up their processing time when scanning images of Mars.

mars-crater-artificial intelligence nasa ai

Image via Adobe Stock

Other than NASA’s research in artificial intelligence (AI) to predict hurricanes and its partnership with Google to make quantum computing accessible, the space agency will also be using AI to classify craters on Mars.

Scientists and researchers from NASA’s Jet Propulsion Laboratory (JPL) will use the new AI tool along with a machine-learning algorithm to speed up their processing time when scanning images.

NASA and JPL’s AI tool

The AI tool will identify unknown craters on Mars. It takes up to 40 minutes to scan a single Context Camera image and the researchers are currently spending hours each day doing that.

According to NASA’s JPL news release, this is the first time that “machine learning has been used to find previously unknown craters on the Red Planet”. Truly a milestone for planetary scientists and AI researchers at NASA’s JPL.

“Typically, scientists spend hours each day studying images captured by NASA’s Mars Reconnaissance Orbiter (MRO), looking for changing surface phenomena like dust devils, avalanches, and shifting dunes. In the orbiter’s 14 years at Mars, scientists have relied on MRO data to find over 1,000 new craters. They’re usually first detected with the spacecraft’s Context Camera, which takes low-resolution images covering hundreds of miles at a time.”

How does it work

The short answer is that it wasn’t easy to train the AI how to discover and identify craters. In order to train the AI took, researchers fed the model more than 6 800 Context Camera images captured by the MRO.

In total, the AI tool is was equipped with 112 000 images, while training of the tool was completed on a supercomputer cluster. Its training isn’t complete and the tool can already “run up to 750 copies of the classifier across the entire cluster simultaneously”.

“Once trained, the classifier was deployed on the Context Camera’s entire repository of about 112 000 images. Running on a supercomputer cluster at JPL made up of dozens of high-performance computers that can operate in concert with one another, a process that takes a human 40 minutes takes the AI tool an average of just five seconds”.

JPL computer scientist Gary Doran explains that it wouldn’t be possible “over 112 000 images in a reasonable amount of time without distributing the work across many computers”.

An exciting symbiosis of human and AI ‘investigators’

That said, JPL computer scientist Kiri Wagstaff explained that the AI tool won’t put scientists out of work. Despite all that computing power, the AI tool still requires a human to check its work. Wagstaff added:

“AI can’t do the kind of skilled analysis a scientist can. But tools like this new algorithm can be their assistants. This paves the way for an exciting symbiosis of human and AI ‘investigators’ working together to accelerate scientific discovery.”

Scientists at JPL and Brown University are hopeful the new tool “could offer a more complete picture of how often meteors strike Mars and also reveal small impacts in areas where they haven’t been discovered before”.

The more craters that are found, the more scientists add to the body of knowledge of the size, shape, and frequency of meteor impacts on Mars. They explain that there are likely many more impacts that they haven’t found yet.

Also read: Nokia wins contract to install first wireless 4G network on the Moon