Astronomy & Space

Scientists employ artificial intelligence to accelerate planetary research.

Future space missions may make use of a novel method for finding planetary craters that will enable researchers to precisely map the surfaces of planets using a variety of data sources.

A group of scientists from the College of Aberdeen has fostered another all inclusive cavity recognition calculation (CDA) utilizing META simulated intelligence’s Fragment Anything Model (SAM).

A brand-new artificial intelligence model known as SAM was unveiled earlier this month. It is capable of automatically “cutting out” any object from any image.

The technology has allowed the team to map craters automatically rather than manually, which takes a lot of time. At the same time, surface characterization that is both more precise and adaptable is made possible by making use of various kinds of data.

“Crater detection is a critical task in planetary science because it allows us to better understand the geology, history, and evolution of celestial bodies like Mars, the Moon, and other planets.”

Dr. Iraklis Giannakis, from the University’s School of Geosciences,

The CDA method has the potential to be a universal solution for detecting craters on a variety of planet surfaces because it can work with a variety of data and celestial bodies.

It could also be used for automatic navigation based on observations of the terrain and assist in determining potential landing sites for human or robotic missions.

In collaboration with colleagues from the university, Dr. Iraklis Giannakis from the School of Geosciences led the research. The results have been published as a preprint on arXiv.

Credit: University of Aberdeen

Dr. Giannakis said, “Hole location is a urgent undertaking in planetary science empowering us to all the more likely grasp the geography, history, and development of divine bodies like Mars, the Moon, and different planets.”

“The need for manual identification is reduced by our universal CDA approach, which takes advantage of the power of SAM to automatically detect craters with high accuracy and efficiency.”

“SAM has proven to be a game-changer for CDA, allowing us to accurately identify craters of various sizes, shapes, and orientations—even in challenging terrain conditions” thanks to its advanced segmentation capabilities.

According to Dr. Giannakis, the development of the CDA has opened up new opportunities for future exploration missions and planetary science.

“Via consequently planning pits researchers can concentrate on their dissemination, size, and morphology to more readily figure out the planetary surface and its advancement over the long haul. This can help in uncovering the geographical history, surface cycles, and possible livability of a planet or moon.”

“Holes can likewise be possible wellsprings of significant assets, for example, water ice on planetary bodies like the Moon or Mars. Scientists can identify potential locations where resources may be concentrated by automatically mapping craters. This can be crucial for planning resource utilization strategies in space exploration scenarios and for future human missions.”

More information: Iraklis Giannakis et al, Deep learning universal crater detection using Segment Anything Model (SAM), arXiv (2023). DOI: 10.48550/arxiv.2304.07764

Topic : News