AI Has Changed the Way We Explore Our Solar System

"Last week at the 2022 American Geophysical Union (AGU) Fall Meeting, planetary scientists and astronomers discussed how new machine-learning techniques are changing the way we learn about our solar system," reports Space.com, "from planning for future mission landings on Jupiter's icy moon Europa to identifying volcanoes on tiny Mercury...." For many tasks in astronomy, it can take humans months, years or even decades of effort to sift through all the necessary data... "You can find up to 10,000, hundreds of thousands of boulders, and it's very time consuming," Nils Prieur, a planetary scientist at Stanford University in California said during his talk at AGU. Prieur's new machine-learning algorithm can detect boulders across the whole moon in only 30 minutes. It's important to know where these large chunks of rock are to make sure new missions can land safely at their destinations. Boulders are also useful for geology, providing clues to how impacts break up the rocks around them to create craters.

Computers can identify a number of other planetary phenomena, too: explosive volcanoes on Mercury, vortexes in Jupiter's thick atmosphere and craters on the moon, to name a few. During the conference, planetary scientist Ethan Duncan, from NASA's Goddard Space Flight Center in Maryland, demonstrated how machine learning can identify not chunks of rock, but chunks of ice on Jupiter's icy moon Europa. The so-called chaos terrain is a messy-looking swath of Europa's surface, with bright ice chunks strewn about a darker background. With its underground ocean, Europa is a prime target for astronomers interested in alien life, and mapping these ice chunks will be key to planning future missions.

Upcoming missions could also incorporate artificial intelligence as part of the team, using this tech to empower probes to make real-time responses to hazards and even land autonomously. Landing is a notorious challenge for spacecraft, and always one of the most dangerous times of a mission.

This is the golden age of AI

[dvice]

Astronomy, medicine, engineering, architecture, biology, chemistry, math, fish breeding, cucumber farming, law, games, music, art, food, logistics, etc. This is the golden age of AI. We are using more and more of it, everywhere. It means more businesses and jobs around AI, which means more knowledge, research, money and innovation around AI, which increases businesses and jobs around AI. And we are not even talking about general or strong AI, just puny little weak AI that can count fishes or stars and not even 100% accuracy, but still better and cheaper than humans.

Re:This is the golden age of AI

[Jzanu]

Precisely. The more of the sensing problems are solved and the more methods are developed to solve language and vision problems the faster our civilization will improve. This will likely be the premise of another scientific revolution. Never underestimate how much science is built on the backs of grad students watching video of mice and ants run around for thousands of hours. And other mind-numbing tasks for which both attention and knowledge are required.

But is this really AI?

[oogoliegoogolie]

Or just big-data processing to find the proverbial needles in the haystacks?

Re:

[dvice]

Difference with big data processing and AI is that for the data processing, you write the rules, most commonly as a search query, and then you search from a big data set for things you need. In AI you don't need to tell the search parameters. You just give sample inputs and expected outputs. Then you need sample inputs one by one and check the results. The major difference is that you don't need to know or even believe that there are search parameters and the AI can still figure them out.

Re:

[accomplishsleuth]

moto x3m [moto-x3m.io]

Life imitating art

[mooterSkooter]

Just don't let it be in control of the pod-bay doors

Re:Life imitating art

[ShanghaiBill]

Just don't let it be in control of the pod-bay doors

That is not specific to AI. Software should not control critical safety systems. They should use mechanical interlocks.

We learned this lesson from Therac-25 [wikipedia.org].

Re:

[r2kordmaa]

Nowadays almost all safety systems are actually controlled by software, just not any random software on any random hardware, there are special methods to get those sorts of things done. Done right, a software based controller is actually safer than a pure hardware analogue for the simple reason that you can have more diagnostic coverage, a software based system can detect and fail safe on more failure modes.

Re:

[careysub]

The real problem with Therac-25 is the code was just a big steaming pile of... spaghetti. The UI code, which simply interacted with the user to get input values, was intermixed with beam control functions. Even by the standards and knowledge of the day (1982) this was know to be bad practice.

Opinion

[Arborl]

I think it's only a matter of time before the computer completely replaces the man in these areas. It's been a couple of years since the Precalculus-type textbooks I found on https://plainmath.net/textbooks/math/precalculus [plainmath.net] have become less popular and more and more use off-the-shelf calculators for everything. And I think that's not a bad thing considering that's how all progress is moving at the moment it's just a little sad.