It seems that currently, artificial intelligence is learning to recognize some patterns that are related to life-supporting conditions on other planets. The final purpose is for AI to be able to predict whether or not life can exist on these extraterrestrial worlds. At least this is what a team of scientists at the Centre for Robotics and Neural Systems at Plymouth University is saying. They are actually presenting their work today at the European Week of Astronomy and Space Science in Liverpool.
According to Christopher Bishop, the leader of the research, the idea is to see how AI would be able to identify potential life forms that exist on other planets and then classify these planets. However, this cannot be done without training. So, the team of experts has been training artificial neural networks to classify five different types of planets in accordance with their life-supporting features. There is the Earth today, the early Earth, mars, Venus and finally, Titan, Saturn’s famous moon. It’s worth mentioning that an artificial neural network is actually an interconnected computing system made out of nodes. These are artificial neurons. The main purpose of such a system is to imitate as closely as possible biological neural networks.
AI is being trained to find alien life on other planets
So, in order to make his network learn more about alien life, Bishop used NASA’s Planetary Spectrum Generator. His network received over a hundred high-resolution spectral profiles of these five planets. They all contained a large number of parameters that had to do with habitability. Angelo Cangelosi, who is the project supervisor, says that the prospects are encouraging for now.
The method can be useful in categorizing certain exoplanets and then used to target these planets in search for potential alien life. And AI plays a key role in all of this.
Image source: pixabay