Name: NASA Is Training an AI to Detect Fresh Craters on Mars
Brand: wired.com
SKU: wired-com-15
Availability: OnlineOnly

critic

NASA Is Training an AI to Detect Fresh Craters on Mars

preview

End UserSectorSource DataTechnology 15 years, NASA’s has been doing laps around the Red Planet studying its climate and geology. Each day, the orbiter sends back a treasure trove of images and other sensor data that NASA scientists have used to scout for safe and to understand the on the planet. Of particular interest to scientists are the orbiter’s crater photos, which can provide a window into the planet’s deep history. NASA engineers are still working on a ; without the rocks that will help them calibrate remote satellite data with conditions on the surface, they must do a lot of educated guesswork when it comes to determining each crater’s age and composition.For now, they need other ways to tease out that information. One tried and true method is to extrapolate the age of the oldest craters from the characteristics of the planet’s newest ones. Since scientists can know the age of some recent impact sites within a few years—or even weeks—they can use them as a baseline to determine the age and composition of much older craters. The problem is finding them. Combing through a planet’s worth of image data looking for the telltale signs of fresh impact is tedious work, but it’s exactly the sort of problem that an AI was made to solve.Late last year, researchers at NASA used a machine-learning algorithm to discover fresh Martian craters for the first time. The AI discovered dozens of them hiding in image data from the Mars Reconnaissance Orbiter and revealed a promising new way to study planets throughout our solar system. “From a science perspective, that’s exciting because it’s increasing our knowledge of those features,” says Kiri Wagstaff, a computer scientist at NASA’s Jet Propulsion Laboratory and one of the leaders of the research team. “The data was there all the time, it’s just that we hadn’t seen it ourselves.”The Mars Reconnaissance Orbiter carries three cameras, but Wagstaff and her colleagues trained their AI using images from just the Context and HiRISE imagers. Context is a relatively low-resolution grayscale camera, whereas HiRISE uses the largest reflecting telescope ever sent into deep space to produce images with resolutions about three times higher than the images used on Google Maps.First, the AI was fed nearly 7,000 orbiter photos of Mars—some with previously discovered craters and others without any—to teach the algorithm how to detect a fresh strike. After the classifier was able to accurately detect craters in the training set, Wagstaff and her team loaded the algorithm onto a supercomputer at the Jet Propulsion Laboratory and used it to comb through a database of more than 112,000 images from the orbiter.“There’s nothing new with the underlying machine-learning technology,” says Wagstaff. “We used a pretty standard convolutional network to analyze the image data, but being able to apply it at scale is still a challenge. That was one of the things we had to wrestle with here.”Most PopularMost recent craters on Mars are small and might only be a few feet across, which means that they appear as dark pixelated blotches on Context images. If the algorithm compares the image of the candidate crater with an earlier photo from the same area and finds that it’s missing the dark patch, there’s a good chance that it’s found a new crater. The date of the earlier image also helps establish the timeline for when the impact happened.Once the AI had identified some promising candidates, NASA researchers were able to do some follow-up observations with the orbiter’s high-resolution camera to confirm that the craters actually existed. Last August, the team got its first confirmation when the orbiter photographed a cluster of craters that had been identified by the algorithm. It was the first time that an AI had discovered a crater on another planet. “There was no guarantee there would be new things,” says Wagstaff. “But there were a lot of them, and one of our big questions is, what makes them harder to find?”The new process could dramatically accelerate crater discovery on Mars and other planets. For the past 15 years, NASA scientists working on the Mars Reconnaissance Orbiter program have had to scour the spacecraft’s images manually, which could take three-quarters of an hour for a single image. “They’ve trained themselves to recognize what these new craters look like, but it can be quite time consuming,” says Ingrid Daubar, a planetary scientist at Brown University who collaborated on the research. The new AI, by contrast, can spot whether an image contains a new dark patch in a mere five seconds.Other than helping determine the age of the Martian surface, Daubar says that craters can also teach scientists a lot about what is just beneath it. For example, about a decade ago, the Mars Reconnaissance Orbiter a new crater that exposed some subsurface water ice. By studying the exposed ice—and how it disappeared over time—scientists were able to get a better sense of how ice is distributed across the surface of the entire planet. But it was a serendipitous discovery. Daubar hopes that an AI that routinely scrutinizes images for hints of new craters, and can alert scientists to them within days or weeks of their formation, will teach us more about Martian history.“The possibility of using machine learning to really delve into large data sets and find things that we otherwise wouldn’t have found is really exciting,” says Daubar. “This particular project identified 60 or 70 new craters we hadn’t seen before. But this is just beginning. We’re looking forward to finding a lot more.”In the future, Wagstaff and her colleagues at the Jet Propulsion Laboratory expect that this type of machine learning will be done in space, to speed up the process even further. Rather than sending all the images back to Earth for processing on giant supercomputers, spacecraft like the Mars Reconnaissance Orbiter will be able to crunch their own data. This will allow for more flexible and responsive missions, since the orbiter won’t have to wait for humans to tell it to check out a point of interest. If it detects a possible crater, it can immediately do a follow-up observation with a more sensitive instrument. And since Mars orbiters are starved for bandwidth, it will also help conserve this precious resource by only sending back images that show interesting changes on the surface.Most PopularFor now, though, that remains a distant goal. The crater work was part of a larger program at NASA called , which aims to implement image change-detection algorithms on orbiters themselves. While detecting changes in images is a well understood problem in AI research, . In their recent work on Mars craters, Wagstaff and the team used 75 cores in a large supercomputer, which is orders of magnitude more computational power than is available to a Mars orbiter.“If you want to do computation on board, currently you’re very limited in what kind of processors you have available,” says Wagstaff. “You don’t have a big supercomputer. You don’t even have a multicore processor. So anything you put up there has to be very, very computationally efficient to achieve change detection.”Integrating AI into future spacecraft is only going to become more important. As technology improves and data transmission rates increase, NASA researchers will have to contend with an ever growing deluge of information. But “needle-in-the-haystack” problems, where the solution is hidden in a massive search space, are exactly the type of challenges machine learning was designed to solve. And once we have AI-powered explorers roaming the solar system, who knows what we might find?More Great WIRED Stories is a staff writer at WIRED, where he covers space exploration and the future of energy. He is the author of (MIT Press, 2019) and was previously the news editor at Motherboard.

End UserSectorSource DataTechnology 15 years, NASA’s has been doing laps around the Red Planet studying its climate and geology. Each day, the orbiter sends back a treasure trove of images and other sensor data that NASA scientists have used to scout for safe and to understand the on the planet. Of particular interest to scientists are the orbiter’s crater photos, which can provide a window into the planet’s deep history. NASA engineers are still working on a ; without the rocks that will help them calibrate remote satellite data with conditions on the surface, they must do a lot of...

public

Daniel Oberhaus

Where tomorrow is realized || Sign up for our newsletters: http://WIRED.com/newsletter

NASA Is Training an AI to Detect Fresh Craters on Mars

No article scores yet.

CRITIC REVIEWS

PUBLIC REVIEWS