According to scholars of NASA’s Goddard Space Flight Center in Greenbelt, Maryland, engineers, and scientists could benefit from machine learning technology. For instance, a computer, trained using advanced neural networks, can identify the potentially fake purchase and initiate an inquiry.

“The benefits are many and the applications are wide-ranging,” Jacqueline Le Moigne, Goddard Senior Fellow and Assistant Chief for Technology, said.

“Scientists could use machine learning to analyze the petabytes of data NASA has already collected over the years, extracting new patterns and new correlations and eventually leading to new scientific discoveries,” she said. “It could also help us monitor the health of a spacecraft, avoid and recover from catastrophic failures, and prevent collisions. It could even assist engineers, providing a wide range of knowledge about past missions—information they would need in designing new missions.”

Goddard scientists and engineers, funded by many NASA research programs, are either researching some of these applications themselves or have collaborated with private industry and academia. They are focusing on a wide range of projects ranging from how machine learning or neural networks could predict real-time crop or pinpoint floods and wildfires to recognize glitches in instruments, and even appropriate robotic landing sites.

“People hear artificial intelligence and their minds instantly go to science fiction with machines taking over, but really it’s just another tool in our data-analysis toolbox and definitely one we shouldn’t neglect because of preconceived notions,” James MacKinnon, Computer Engineer at Goddard, said.

James MacKinnon has trained algorithms to accurately recognize wildfires using remote-sensing images.

“The key here is processing the data onboard, not only for wildfires but for floods. There are a lot of things you could do with this capability,” he said. He is also creating machine-learning methods to recognize single-event upsets in “spaceborne electronic devices”.

Matt McGill, an expert in lidar techniques, is collaborating with Slingshot Aerospace. He specifically wants to see if machine learning can filter out the common sound production in lidar measurements. Current noise culling techniques take a lot of time.

“The idea is that algorithms, once trained, can recognize signals in hours rather than days,” McGill said.

Ron Zellar, Goddard engineer, and Antti Pulkkinen, Goddard heliophysicist, are inquiring whether the solar storms are responsible for dolphin’s stranding.

“We can’t assume a causal relationship,” said Zellar. “That’s what we’re trying to find.” The team is funded by Goddard Fellows Innovation Challenge and they are using machine learning to deeply study and investigate the environmental data to check if they can demonstrate a cause.

A team consisting of Dante Lauretta, a professor at the University of Arizona, Goddard scientists, and Chris Adami, a machine-learning expert at Michigan State University, are funded by NASA and are inquiring the potential of networked algorithms. Their aim is to train the sensors onboard to study images and find out the asteroid’s features and shapes.

“The point is to cut the computational umbilical cord back to Earth,” Bill Cutlip, Senior manager at Goddard and a team member, said. “What we’re trying to do is train an algorithm to understand what it’s seeing, mimicking how the human brain processes information.”