This article is more than 1 year old

SETI: How AI-boosted satellites, robots could help search for life on other planets

Software scans to give rovers a clue

AI-infused algorithms developed to find signs of life in extreme terrestrial environments could help robotic rovers sent to other planets search for signs of alien life, scientists suggested in new research published in Nature Astronomy on Monday.

Microbes living in extreme environments are normally constrained to specific areas that have the right conditions and enough resources for them to survive. They leave traces of their existence in the form of strange patterns and grooves that can be detected by analysing surface features.

A large team of researchers led by the SETI Institute believe AI algorithms can be trained to detect these biosignatures by analyzing surface maps obtained from spacecraft orbiting a planet. Rovers equipped with the computer vision software could then be sent to look for these signals of extraterrestrial life.

The authors' thesis is based on analysis of terrain around the Salar de Pajonales, an ancient lakebed in Chile that is home to communities of photosynthetic bacteria. The region is 3,500m above sea level and the microbes therefore live with high levels of ultraviolet radiation and salinity. One of the signs of their struggle to survive is fractal-like patterns and ridges on the lakebed. 

The research team trained a convolutional neural network on those images and were able to detect other regions where these organisms lurked, with up to 87.5 per cent accuracy. Rovers running such AI software could therefore hunt for similar biosignatures more efficiently than using current methods of landing somewhere promising and searching based on satellite snaps alone, the researchers argued. 

"Our framework allows us to combine the power of statistical ecology with machine learning to discover and predict the patterns and rules by which nature survives and distributes itself in the harshest landscapes on Earth," Kim Warren-Rhodes, first author of the study and a SETI Institute Senior Research Scientist, said in a statement. 

"We hope other astrobiology teams adapt our approach to mapping other habitable environments and biosignatures. With these models, we can design tailor-made roadmaps and algorithms to guide rovers to places with the highest probability of harboring past or present life—no matter how hidden or rare."

Over 7,765 images of the Salar de Pajonales collected from drone footage and 1,154 samples directly taken from the lakebed detecting microbes in the salt domes, rocks, and crystals were used to train the model. The software confirmed that these photosynthetic bacteria were concentrated in small areas that were near water sources.

If future rovers are to find biosignatures using similar AI algorithms, scientists and engineers will have to work with multiple datasets obtained using different instruments in space and on the ground, Nathalie Cabrol, co-author of the paper and the principal investigator of the SETI Institute NASA Astrobiology Institute team, said.

"While the high-rate of biosignature detection is a central result of this study, no less important is that it successfully integrated datasets at vastly different resolutions from orbit to the ground, and finally tied regional orbital data with microbial habitats,"

"With it, our team demonstrated a pathway that enables the transition from the scales and resolutions required to characterize habitability to those that can help us find life. In that strategy, drones were essential, but so was the implementation of microbial ecology field investigations that require extended periods (up to weeks) of in situ (and in place) mapping in small areas, a strategy that was critical to characterize local environmental patterns favorable to life niches."

In other words, the technique will be effective if there is already substantial evidence for alien microbial life and if a planet's environment and surface have been mapped carefully. Rovers will only be able to find biosignatures more effectively if they already know what to look for. 

The Register has asked the SETI Institute for further comment. ®

More about


Send us news

Other stories you might like