Published in


NASA is Using Intel’s Deep Learning to Find Better Landing Sites on the Moon

People have already seen tons of machine learning and neural network applications, but recently NASA announced that their Frontier Development Laboratory (FDL) would cooperate with Intel to apply deep learning for space exploration. One of the recent accomplishments is the application to build a moon map. Scientists at Intel obtained about 200 TB of data from NASA, and then drew the three-dimensional images of the lunar poles by applying deep learning. With the help of deep learning, scientists have enhanced the possibility to restore and locate the craters as well as the details on the moon.

Naveen Rao, Intel’s General Manager for the Artificial Intelligence Products Group, believes that the industry has more availability and access to apply the latest and greatest computational tools than the research world. It is significant for scientists to learn and catch up to those fresh and advanced methods, which are able to reduce computational cost and at the same time make more sense to scientists, because these days, it is much more easier to gather data than to make sense of or to interpret it. Rao acclaimed that the Nervana system was able to accomplish in about a minute the same amount of tasks which usually took scientists two or three hours to finish, and its results showed a high accuracy of about 98.4%.

After data interpretation, the result is returned to NASA. NASA is then able to apply this result to better determine a landing point on the lunar surface for lunar rovers of the future. Why is this important for NASA? Because the optimized landing point will decide if the lunar rovers can have better solar exposure to power on-board rovers and save on total cost.

However, this is not the only project that FDL focuses on. Unlike other branch laboratories in NASA, FDL likes applying the latest artificial intelligence technologies to their hot topics, such as Space Weather, Space Resources, and Planetary Defense. Therefore, machine learning is extremely useful and well-fitted to these projects.

Besides the reconstruction of the moon maps, I’d like to introduce another feature of this project, the Lunar Water and Volatiles. The objective here is to pinpoint the locations of water and other volatile resources on the moon. Those volatile resources can be hydrogen, carbon dioxide, nitrogen etc., which are key elements in producing fresh air supply, rocket fuels, and other significant materials. All of the aforementioned materials have to be produced and loaded enough, so that people are able to achieve space exploration beyond Mars. “Space data is often massive, multi-dimensional and dynamic.” said by James Parr, FDL Director. This is why high quality and reliable deep learning is utilized here, to quickly learn and analyze space data in order to accelerate the pace of outer space exploration. However, even if researchers successfully find the resources with the help of deep learning methods, they still need to plan a route to deploy the missions. Here comes to the importance of the landing point mentioned at the beginning of this article. In order to avoid lunar hazards such as craters, boulders, and cliffs, an automated map needs to be built to identify those hazards, and provide better landing locations for lunar rovers.

Figure 1. Intel and NASA Researchers will be applying AI to locating water and volatile resources on the moon. (Photo Credit: James Parr)

On August 17, Intel hosted the FDL Wrap-Up Event at Santa Clara campus and shared their result for the space research. With the successful cooperation between Intel and NASA, future space exploration and aerospace engineering will become interdisciplinary fields, combining traditional engineering and deep learning.

Figure 2. Planetary scientist Eleni Bohacek speaks Thursday at the NASA Frontier Development Lab (FDL) Wrap-Up Event.


[1]Intel Showcases Application of AI for Space Research at NASA FDL Event.
[2]Prospecting for Space Resources with Intel® Nervana™.

Blog Author: Brian Heater

Author: Bin Liu | Editor: Haojin Yang | Localized by Synced Global Team: Xiang Chen




We produce professional, authoritative, and thought-provoking content relating to artificial intelligence, machine intelligence, emerging technologies and industrial insights.

Recommended from Medium

A Bright Beaver Moon Mangles Meteors, Tiny Mercury Transits the Sun, and Vesta Veers Closer!

Milkway exchange coming

February Full Moon: Snow Moon 2022

Enjoy Asteroid Juno, Mercury at Dusk, and Evening Gas Giants, plus a Jeweled Scimitar on the Moon!

Testing the first electrostatic radiation shield in deep space

Moon and Mars from dusk to dawn (and even more)

Stargazing Tonight: The Planets and Fading NEOWISE

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


AI Technology & Industry Review — | Newsletter: | Share My Research | Twitter: @Synced_Global

More from Medium

Inside Meta’s New Architecture for Build AI Agents that Can Reason Like Humans and Animals

Cash App Labs Modifies the Very Deep VAE to Achieve a 2.6x Speedup and 20x Memory Reduction

Multimodal AI Modeling is the Future, But It’s Also Pandora’s Box

How Microsoft & OpenAI Are Squeezing the Best Out of GPT-3