BEIJING, Dec. 27 (Xinhua) — The use of artificial intelligence techniques such as machine learning to analyze future survey data from the James Webb Space Telescope or the Lingsun Exoplanet Survey Satellite (TESS) will help astronomers search for extraterrestrial life, according to a NASA statement. Detect the Earth’s proximity to asteroids.
Using artificial intelligence techniques such as machine learning to analyze future survey data from the James Webb Space Telescope or the Exoplanet Survey Satellite (TESS) will help astronomers search for extraterrestrial life and probe earth’s proximity to asteroids.
Giada Arney, an astrobiologist at the Goddard Space Flight Center, said aia technology is very important, especially for large data sets in the field of exoplanets, because the data we get from future observations is sparse and noisy, which is really hard to understand. So using these tools has great potential to help us.
NASA has partnered with Intel, IBM, Google and others to develop advanced machine learning technologies, and every summer, NASA brings in scientists and astronomers for an eight-week workshop called the Frontier Development Laboratory (FDL).
Shawn Domagal-Goldman, an astrobiologist at the Goddard Space Flight Center, said: “The Frontier Development Lab is like a first-class musician, with different instruments, gathering and improvising in the garage. “
In 2018, Gorman and Arnie led an FDL team that developed a machine learning technique that uses brain-like “neural networks” to analyze space images and identify their chemical properties based on light waves released or absorbed by molecules in the atmosphere of exoplanets, similar to nerve cells in the human brain. Ability to process and transmit information.
The researchers used this neural network technique to more accurately identify atmospheric molecular diversity of exoplanet WASP-12b than traditional methods, and in addition, it can also identify when there is insufficient data, which would be important if we adopted these predictions.
The researchers note that while neural network technology is still in the development stage, it could one day be used to study data collected by telescopes, which could help narrow the range of exoplanet candidates that deserve further study.
Other FDL technologies have also been well used, such as a 2017 machine learning program that creates 3D models of asteroids in just four days, including their size, shape and rotation speed. At the same time, such artificial intelligence programs are particularly important for detecting and shifting potential asteroids that threaten The Earth.
NASA collects about 2 gigabytes of data every 15 minutes from space probes, says sun physicist Madhulika Guhathakurta, which is why more tools are needed for research and analysis.
In addition, the researchers suggest that artificial intelligence technology will be used in future spacecraft to make it easier for the spacecraft’s systems to make real-time scientific analysis, thus saving the time it takes to communicate with scientists on the ground.
Still, Arnie points out, artificial intelligence technology won’t replace humans any time soon, because we also need to examine the final results and draw conclusions based on expert research and analysis.