The fascination with extraterrestrial life has been a long-standing subject for humans, as evidenced by the countless books, movies, and TV series dedicated to the topic. While the search for life beyond our planet has progressed considerably in recent years, it remains in its early stages. Despite sending rovers to Mars, remote exploration has yielded low accuracy in finding extraterrestrial life.
To improve the accuracy of this search, the Search for Extraterrestrial Intelligence (SETI) Institute is exploring the use of artificial intelligence (AI) to detect microbial life in the depths of icy oceans on other planets. In a recent study published in Nature Astronomy, the team at SETI details how they trained a machine-learning model using a convolutional neural network (CNN) algorithm to scan data for signs of microbial life or other unusual features that could be indicative of alien life.
The team, led by Kim Warren-Rhodes, used statistical ecology and AI-detected biosignatures to develop a system with up to 87.5% accuracy, compared to only 10% for random searches. The system can potentially reduce the search area by up to 97%, making it easier for scientists to locate potential chemical traces of life. To test the system, the team focused on sparse lifeforms that dwell in salt domes, rocks, and crystals at Salar de Pajonales on the boundary of the Chilean Atacama Desert and Altiplano.
Warren-Rhodes and his team collected over 8,000 images and 1,000 samples to search for photosynthetic microbes that may represent a biosignature on NASA’s ‘ladder of life detection’ for finding life beyond Earth. The team also used drone imagery to simulate Martian terrain images to examine the region. They found that microbial life in the region is concentrated in biological hotspots that strongly relate to the availability of water.
The machine learning tools developed can be used in robotic planetary missions like NASA’s Perseverance Rover. The tools can guide rovers towards areas with a higher probability of having traces of alien life, even if they are rare or hidden. With these models, we can design tailor-made roadmaps and algorithms to guide rovers to places with the highest probability of harbouring past or present life.
Post Your Comments