Ever wondered about the power of Artificial Intelligence and Machine learning? It has so much potential. The recent artificial intelligence model – Spock AI might prove to be a remarkable achievement in Astrophysics and space sciences.
Apart from our solar system, there are many other multi-planetary systems in the universe identified by the world of space science. We have 701 known multi-planetary systems as of now, out of which 280 have two exoplanets and many other systems contain a large number of planets. Scientists have very little knowledge about the orbits of those planets right now.
The SPOCK Artificial Intelligence Model would make it possible to predict the future of the planets within multi-planetary systems.
Agroup of NASA and Princeton scientists proposed their new Artificial intelligence model named SPOCK- Stability of Orbital Configurations Classifier. This AI model is remarkably fast in calculating the configuration of the orbital stability of planets. This means SPOCK will calculate the configurations of planetary motions for collisions and stability 100,000 times faster than the current duration. At present, the scientists have to perform robust calculations using supercomputers which takes years to find a stable configuration. The machine learning ability of the SPOCK Artificial intelligence model could complete this task in minutes. The paper is published in ‘Proceedings of the national academy of science’ journal. Read Spock Paper PDF here.
This is how the SPOCK Artificial Intelligence works.
Prediction is carried out by simulating 10,000 orbits and calculating 10 summary metrics of the planetary system’s dynamics by the scientists. After that, the SPOCK A.I. model calculates which of these 10 features would have a substantial future if it continued to calculate further configurations until the planets orbit around their star for about 1 billion times. “We can’t categorically say ‘This system will be OK, but that one will blow up soon,” says Daniel Tamayo, one of the contributing scientists to the A.I. model. “The goal instead is, for a given system, to rule out all the unstable possibilities that would have already collided and couldn’t exist at the present day.” He explains it with his pictures.
These two pictures are of Kepler – 431 planetary system involving three planets. There is very little knowledge about their orbits right now. On the left is the picture taken after traditional computing and on the right is the picture taken after SPOCK calculations.In this picture, there are too many superimposed orbits for each planet consistent with observation. Achieved by rigorous calculations over Supercomputers.
The SPOCK Artificial intelligence model has removed the unstable configurations that would have already resulted in collisions.
As we can see, there are fewer orbital configurations in this picture above. These are the stable configurations achieved by Spock. By doing this with previously established methods, it would take more than a year to establish stable orbital configurations. With SPOCK, it takes just 14 minutes!
Isn’t this outstanding? SPOCK would help a lot in the life prediction of the planetary systems and the organization of the planetary positions and their motion. Many new findings of our universe in the world of Astrophysics and other Space sciences would now come up faster than ever before. According to Daniel Tamayo, their Artificial intelligence model SPOCK would ultimately determine which planets are going to “live long and prosper”. Here is Daniel Tamayo’s official tweet about SPOCK.
Daniel and his colleagues named SPOCK after the fictional star trek character named Spock, who is a Vulcan, and the phrase “live long and prosper” is a Vulcan salutation.
