ML model for habitable planets

ML model for habitable planets
ML model for habitable planets
ML model for habitable planets

Pakyong , 27 Marcha : Artificial intelligence may aid in the search for life on Mars and other extraterrestrial planets

A recently developed model using machine learning techniques could guide scientists towards the most hopeful habitable locations beyond our planet.
A recently created machine learning tool has the potential to aid scientists in their quest to detect signs of life on Mars and other extraterrestrial planets.Scientists face significant constraints in obtaining samples from other planets, which means they have to rely on remote sensing techniques to identify possible indications of extraterrestrial life. Consequently, any approach that can assist in directing or refining this search would be immensely valuable.

In light of this, a group of multidisciplinary scientists, headed by Kim Warren-Rhodes from the SETI Institute in California, created a map of the sparse organisms inhabiting salt domes, rocks, and crystals found in the Salar de Pajonales, a salt flat located on the border of the Chilean Atacama Desert and Altiplano high plateau.
Warren-Rhodes collaborated with Michael Phillips from Johns Hopkins University’s Applied Physics Laboratory, and Freddie Kalaitzis, a researcher from the University of Oxford, to develop a machine learning algorithm capable of identifying the patterns and rules governing the distribution of life across the inhospitable region. Through this training, the model was taught to recognize these patterns and rules in a variety of landscapes, including those that could exist on other planets.
By merging statistical ecology with AI, the team found that their system could identify and recognize biosignatures up to 87.5% of the time. This is a remarkable improvement compared to the mere 10% success rate achieved through random searches. Additionally, the program could significantly reduce the search area by up to 97%, which could assist scientists in narrowing down their search for possible chemical indicators of life, also known as biosignatures.

“Our methodology enables us to merge the capabilities of statistical ecology with machine learning to identify and forecast the patterns and rules that enable nature to survive and propagate in the most challenging environments on Earth,” stated Warren-Rhodes. “We expect that other astrobiology teams will adopt our approach to map other habitable environments and biosignatures.”

The researchers claim that such machine learning tools could be used for robotic planetary missions like NASA’s Perseverance rover, which is currently searching for signs of life in the Jezero Crater on Mars.
The group opted to use Salar de Pajonales as a testing ground for their machine learning model because it resembles the dry and arid terrain found on Mars today. The area is a dry, high-altitude salt lakebed that is exposed to high levels of ultraviolet radiation. Despite being recognized as a severely unwelcoming habitat for living organisms, Salar de Pajonales still sustains some forms of life.

For breaking news and live news updates, like us on Facebook or follow us on Twitter and Instagram Visit

The Voice Of Sikkim | Sikkim Live | Himdarpan | The Siliguri Today | Samvad