A team of scientists at NASA Goddard Space Flight Centre (GSFC) has designed a new AI system that can decide which data it can transmit home.

Since its initiation, AI has captured the attention of the world owing to its wide range of capabilities. Even NASA (National Aeronautics and Space Administration) has been planning to enlist AI for future space exploration and other programs. Recently in June 2020, the American space agency announced that it has been training the system of AI that will aid scientists in their quest to look for signs of ancient life on Mars and other planets and moons. The program will be spearheaded by the European Space Agency (ESA) Rosalind Franklin 'ExoMars' rover mission. It will be heading for the red planet in 2022/23, before moving beyond to moons such as Jupiter's Europa, and of Saturn's Enceladus and Titan.

This news comes after a team of scientists from the NASA Goddard Space Flight Centre (GSFC) have announced the first results from new AI systems, which are projected to be installed in space probes. These systems are capable of identifying geochemical signatures of life from rock samples. The exciting feature of these intelligent systems is their ability to choose between what to analyze and what to transmit back to Earth will overcome severe limits on how information is transmitted over huge distances in the search for life from distant planets. It is these AI systems that will be onboard the ExoMars mission. This will be an upgrade from the existing rovers that cost us money and time by sending all data back to Earth due to their incapability of optimizing those data sets.

Speaking about this extraordinary feat, the lead researcher Victoria Da Poian from NASA's GSFC calls this is a visionary step in space exploration. "It means that over time we'll have moved from the idea that humans are involved with nearly everything in space, to the idea that computers are equipped with intelligent systems, and they are trained to make some decisions and can transmit in priority the most interesting or time-critical information," Da Poian said in a press release.

Da Poian, who is also an aerospace engineer at NASA Goddard, cites that generally, scientists waste hours in their attempt to comprehend and analyze all the data. And projects like ExoMars can help reduce their burden by allowing real-time decisions to happen on-site. This AI project will commence in 3 phases. In the first phase, the AI system understands the data and translates it into a readable format. In the second phase, data is clustered into different groups: interesting data that require the scientists' attention, not so interesting data, and data that is similar to something that the scientists have observed before. When new raw data is received, the software tells the scientists which previous samples match this new data. Lastly, in the third phase, which is currently in development, would be a neural network capable of suggesting similar data that scientists have previously taken note of. This is synonymous with the algorithms used for Netflix recommendations.

Her co-researcher Eric Lyness, also from the GSFC, informed that data from a rover on Mars could cost as much as 100,000 times as much as data on our cell phone, so it is important to make those bits as scientifically valuable as possible. This is because we need smart instruments for planetary exploration. He added, "When first gathered, the data produced by the Mars Organic Molecule Analyzer (MOMA) toaster-sized life-searching instrument will not shout out ''I've found life here', but will give us probabilities which will need to be analyzed." The MOMA analyzer is claimed to be the largest instrument on the Rosalind Franklin rover. It mills the samples, heats them, and performs mass spectrometry and gas chromatography to identify molecules. Besides, the rover is predicted to more likely land at Oxia Planum, near the Martian equator. This area has a smooth landing spot and also has the potential to hold any preserved biosignatures.

"These results will largely tell us about the geochemistry that the instruments find. We're aiming for the system to give scientists directions, for example, our system might say "I've got 91% confidence that this sample corresponds to a real-world sample and I'm 87% sure it is phospholipids, similar to a sample tested on July 24th, 2018 and here is what that data looked like". We'll still need humans to interpret the findings, but the first filter will be the AI system," he continues. Eric is a software lead in the Planetary Environments Lab at GSFC.

Moreover, this is not the first nor only instance AI will be employed in space technologies. The National Astronomical Observatory of Japan (NAOJ), is using AI to classify galaxy morphologies with an accuracy of 97.5% successfully. By using the deep-learning technique, a type of AI, it has identified nearly 560,000 galaxies with spiral patterns in a large dataset of images containing about 80,000 galaxies obtained with the Subaru Telescope. A few months ago, Airbus, the German Aerospace Center DLR, and IBM launched the technology-demonstration Project CIMON—Crew Interactive Mobile Companion—the first AI-powered robot in space. CIMON, which is created using 3D printing technology, is a free-floating, sphere-shaped interactive companion that can assist the astronauts in their daily work. It is capable of hearing and seeing and serves through searching for objects, inventory management, documenting experiments, videography, and photography.

As we are on the cusp of exploring what our solar system and space beyond, looking for life beyond Earth or analyzing surfaces of moons and many more, AI will surely help us take the giant leap for mankind, which Neil Armstrong had quoted during his landing on Moon. This will be a pivotal point and advance our quest in space exploration.