HomeArtificial IntelligenceArtificial Intelligence NewsFinding 60 potential Habitable Planets through AI

Finding 60 potential Habitable Planets through AI

Indian astronomers have developed a new approach to identifying habitable planets with high probability using an artificial intelligence-based algorithm. As a result, out of 5,000 known planets, about 60 potentially habitable new planets were discovered.

According to research by astronomers from the Indian Institute of Astrophysics (IIA), an institute under the Ministry of Science and Technology, under the Government of India, together with astronomers from BITS Pilani, the campus Goa, a new initiative for an anomaly identification procedure has been discovered, whereby all potentially habitable planets can be found with a high probability.

“This method is based on the assumption that the Earth is an anomalous object, with the possibility that there are a few other anomalies among the thousands of data points… There are 60 potentially habitable planets among them. approximately 5000 confirmed planets, and nearly 8,000 planets Proposed candidate planets. Assessment based on their close similarity to Earth.

Searching New Habitable Planets

Speaking on the topic, Dr Snehanshu Saha of BITS Pilani said that Earth is the only habitable planet among many worlds and has been described as an anomalous planet. The researchers investigated whether candidates for similar anomalies could be detected using new anomaly detection methods.

Group IIA also describes the pivot of the theory that habitable (potentially) exoplanets are anomalies around the well-known problem of anomaly detection in predictive maintenance of systems. industrial system. Furthermore, the anomaly detection method suitable for industrial systems is also relevant to the detection of habitable planets because in both cases the anomaly detector processes data “unlikely”. equilibrium”, where anomalies (number of habitable exoplanets or unusual behavior of industrial components) are outliers. These numbers are much less than the standard data.

A Landmark Breakthrough!

Manually scanning thousands of planets to identify planets that may resemble the Earth can be tedious. Artificial intelligence (AI) can be used efficiently to find habitable planets.

The AI-based process, known as the MultiStage Memetic Binary Tree Anomaly Identifier (MSMBTAI), is based on the new Multi-Stage Memetic Algorithm (MSMA). MSMA uses a generic term for memes, knowledge, or ideas that are transmitted from one person to another by imitation. Memes show the evolution of posterity across different cultures, and as a result, can create new learning mechanisms across generations. This algorithm also serves as a rapid screening device for assessing the habitability prospects of the observed properties.

This study identified several planets that share anomalous properties similar to Earth through the proposed technology. This shows pretty good results and is consistent with what astronomers believe. Interestingly, this process yielded similar results in detecting anomalous exploration compared to the time when surface temperature was not used as a feature.

source link

 

Most Popular