Researchers Develop Use AI to Classify Planets


Artificial neural networks (ANNs) are being used to classify planets and determine the possibility of life on other worlds. The AI compares planets to early Earth, present-day Earth, Venus, Mars or Titan, the largest moon of Saturn. The five bodies are actually the most potentially livable objects in the solar system and have a high probability of hosting life. Data gathered by the ANNs could help astronomers plan future exploration missions to planets that may hold life.

The only planet known to support life is Earth, but the space is so vast that there is always a possibility that life exists on other planets. Christopher Bishop, a Centre for Robotics and Neural Systems researcher at Plymouth University, said that they are interested in these artificial neural networks for exploring exoplanet systems at a range.

Artificial neural networks can select and identify patterns in large amounts of data, a task that would be too time-consuming and complex for scientists to do. ANNs imitate the learning process of the human brain. These computer systems are given spectra or atmospheric observations from early Earth predictions, present-day Earth, Mars, Titan and Venus. Life outside of Earth is yet to be discovered, so the artificial neural networks use a probability of life measurement based on the orbital and atmospheric properties of these five bodies.

ANNs can determine the habitability of various planets using these spectral profiles. This can help researchers save a lot of time by letting them focus on the most favorable targets. The use of artificial neural networks has been successful for classifying new planets. Project supervisor Angelo Cangelosi said that with the results so far, this technique may prove to be very beneficial for classifying various kinds of exoplanets using results from near-Earth and ground-based observatories.

Researchers are hoping to use the technique to assess data collected during future missions like the European Space Agency’s Atmospheric Remote-sensing Exoplanet Large-survey (ARIEL) space mission and NASA’s spacecraft James Webb Space Telescope (JWST) that will launch in 2020. The JWST will explore the space to discover the history of the universe and will focus on the birth of stars, planets and protoplanetary systems, first light in the cosmos and galaxy assembly in the early universe. It is also expected to take photographs of celestial objects.

ARIEL is scheduled for launch in 2028 and will observe at least 1,000 known exoplanets through the transit method. The mission will study and characterize the thermal structures and chemical composition of the planets.