Beyond the Unknown: Applications of Artificial Intelligence In Spaceby Preetipadma February 5, 2021
While Artificial intelligence is already being leveraged in many IT sectors, what does it offer in space industry?
Artificial intelligence (AI) is rapidly being explored or adopted by many industries for a wide array of applications. Today it is creating a string of opportunities in space industry use-cases too. As artificial intelligence emerges as a popular theme in space exploration, it is also being deployed for many critical tasks too.
For instance scientists have leveraged artificial intelligence, for charting unmarked galaxies, supernovas, stars, blackholes, and studying cosmic events that would otherwise go unnoticed. One of the recent illustration of this application was when CHIRP (Continuous High-Resolution Image Reconstruction using Patch Priors) Algorithm helped in creating first-ever image of a black hole. CHIRP is a Bayesian algorithm used to perform de-convolution on images created in radio astronomy. It used the image data from the Event Horizon Telescopes to carry further image processing. Even images from the Hubble Space Telescope are used to simulate galaxy formation and further classification using deep learning algorithms.
Discovery of ExoPlanets and Alien Life
Artificial intelligence also proves resourceful in classifying heavenly bodies, especially exoplanets. A couple of years ago, a research team developed an artificial neural networks algorithm, to classify planets, based on whether they resemble present-day Earth, early Earth, Mars, Venus or Saturn’s largest moon, Titan. These five bodies are most potentially habitable objects in our solar system and are therefore associated with a certain probability of life.
In regards to life in outer space, Researchers at NASA’s Frontier Development Lab (FDL) employed generative adversarial networks, or GANs, to create 3.5 million possible permutations of alien life based on signals from Kepler and the European Space Agency’s Gaia telescope.
Besides, NASA has teamed up with Google to train its artificial intelligence algorithms to sift through the data from the Kepler mission to look for signals from an exoplanet crossing in front of its parent star. With the help of Google’s trained model, NASA managed to discover two obscure planets — Kepler-90i and Kepler-80g. In 2019, astronomers from the University of Texas at Austin, teamed with Google, to use AI for uncovering two more hidden planets in the Kepler space telescope archive (Kepler’s extended mission, called K2). They used an AI algorithm that sifts through Kepler’s data to ferret out signals that were missed by traditional planet-hunting methods. This helped them discover the planets K2-293b and K2-294b.
Under the Artificial Intelligence Data Analysis (AIDA) project, which is funded under the European Horizons 2020 framework, an intelligent system is being developed that can read and process data from space. The key object of this project is to enable the discovery of new celestial objects, using data from NASA.
AI applications can also found in the field of satellite imagery. Data based on satellite imagery offers insights on several global-scale economic, social and industrial processes, which was previously not possible. Some examples include Earth Observer 1 (EO-1) satellite, SKICAT, ENVISAT. These satellites leverage artificial intelligence to provide actionable insights for agencies, governments and businesses, and help them in making accurate decisions.
While humans are capable of interpreting, understanding, and analyzing images collected by satellites, it does cost us time and resources while waiting for a satellite to move back around to the same position to further refine image analysis. Artificial intelligence helps eliminate the necessity for large amounts of communication to and from Earth to analyze photos and helps determine whether a new photo needs to be taken. Moreover, it saves processing power, reduces battery usage, and fast-tracks the image gathering process.
In case of space mining, artificial intelligence will augment mining machinery with intelligence that will empower them to extract minerals and identify any hazards or solve minor issues at hand without the need for immediate support from humans on Earth. Meanwhile, NASA is also developing a companion for astronauts aboard the ISS, called Robonaut, which will work alongside the astronauts or take on tasks that are too risky for them. According to NASA’s blog, Robonaut 2 is slowly approaching human dexterity implying tasks like changing out an air filter can be performed without modifications to the existing design.
Artificial intelligence has also helped us develop space humanoids like Kirobo from Japan Aerospace Exploration Agency, Dextre from Canadian Space Agency, and AILA from German Research Center for Artificial Intelligence to help astronauts in space missions. NASA’s free-flying robotic system, Astrobee, uses AI to help astronauts reduce their time on routine duties, leaving them to focus more on the things that only humans can do. We also have CIMON or (Crew Interactive Mobile Companion), an AI powered robot that floats through the zero-gravity environment of the space station to research a database of information about the ISS. In addition to the mechanical tasks assigned, CIMON assesses the moods of its human crewmates at the ISS and interacts accordingly with them.