It’s been thirty years since that moment as it was the first scientific proof of the existence of planets outside the Solar system. The time of publication, the official status of exoplanets got 3767 objects when the total number of candidates more than 4,500.
Most of these planets are very harsh and definitely uninhabitable worlds, but some of them, according to scientists, may possess suitable conditions for its occurrence. At least they are not too hot and not too cold in order to maintain the presence of water on the surface in liquid form. And the water is known to be one of the sources of life.
Of course, the main cause of finding new exoplanets is to find life beyond Earth. Why else spend a lot of money for the construction of new telescopes and new technologies to explore space? Therefore, scientists from Columbia University (USA) has developed a new system which can simplify to “hunt” for potentially habitable worlds. Using machine learning algorithms, researchers have created a technology that allows more efficiently to determine the possibility of survival of one or the other extrasolar planets in a stable orbit.
In this work, the researchers focused on so-called “Tatooine”, or exoplanets orbiting binary stars just like the native desert world of Luke Skywalker from “Star wars.” Formally known in scientific circles as circumbinary planet, they may be subject to enormous orbital changes, since are always in the gravitational pool of two stars. Attracted to one star or the other, they risk eventually to be ejected from the system, and in the worst case, to fall on one of their stars.
Scientists have deduced the equation that helps to determine the long-term stability of the orbit circumbinary planets, however, according to Chris Lama, head of development, which is now in question, this equation may not give accurate data to account for all possible circumstances.
“The problem is that if the system of three or more bodies, the movement becomes “chaotic” as they say in physics and mathematics,” comments Lam.
“So there are borderline cases where the equation predicts that the system is unstable, when it actually is stable, and Vice versa. We felt that to cope with this problem, we will help neural network”.
The ability to predict whether a planet thrown out of its system – not just a fad, this is an additional opportunity to determine the potential habitability of one or another of the world. In the end, for the emergence and development of life, at least that which occurs on Earth, it took several billion years. In other words, it will be no chance if we are talking about a planet wandering in space and not tied to its sun.
For a more effective method of determining survival “Tatooine” Lam and his colleagues created a machine learning algorithm, which scientists are trained using 10 million simulated planets. As noted by Lam, after several hours of experiments, and settings the system was able to surpass the accuracy of traditional of the equation “in all respects”.
Scientists expect that the new space telescope TESS space Agency NASA recently successfully put on orbit, will be able to discover many new circumbinary planets, and the development of researchers from Columbia University, says Lam, can help in the study of these worlds.
“Our model will help astronomers to understand which regions are best suited to search for planets around binary systems. I hope this will help us not only to discover new exoplanets, but also to better understand their characteristics,” — said the scientist.
AI has surpassed astronomers in the efficiency of determining survival of exoplanets
Nikolai Khizhnyak