Prospecting for Space Resources with Intel® Nervana™

Jul 31, 2017

Author Bio Image

Shashi Jain

Innovation Manager, Software and Services Group

Author Bio Image

Katie Fritsch

Sr. Ecosystem and Partnership Manager, Artificial Intelligence Products Group

We live in an exciting time for commercial space exploration. Every week, we see exciting leaps forward: reusable rockets, plans to mine the moon, and missions to colonize Mars. Sound like science fiction? It’s not. A little-known NASA program is there to accelerate the pace of that innovation. The NASA Frontier Development Laboratory (FDL), run by the SETI Institute (yes, the one involved with finding extra-terrestrials) is a public-private partnership chartered with solving problems in space exploration in conjunction with industry partners such as Intel, NVidia, and IBM.

Applied Research

FDL is an applied research accelerator that employs the latest artificial intelligence (AI) technologies to heady topics such as Space Weather, Space Resources, and Planetary Defense.  Machine learning is well suited to these problems, as it is able to quickly infer useful knowledge from large volumes of from data. The program attracts the best and brightest in fields ranging from planetary science to 3D modeling along with post-doc and graduate researchers at the top of artificial intelligence and deep learning.  During the 9-week program, teams of 4-5 researchers work with mentors from industry, academia, and NASA to solve key problems in each area. One team was focused on mapping asteroid shapes from radar imagery, another on predicting the interactions of the sun with earth’s atmosphere.  While it’s not a lot of time, the teams are taking on problems that will help NASA solve complex problems.

Domain experts from all over the world come together at NASA Ames Research Center in California for the duration of FDL. (Photo Credit: James Parr)

How did Intel get involved?

A few months ago Intel was approached by James Parr, Director of NASA FDL project, to join as a sponsor for FDL 2017. James challenged us to be visionary and help do our part towards making the human race a space-faring civilization, just by collaborating with the tools we already have. We were sold immediately.

Intel envisions AI as one of the next big waves of computing that will transform how businesses operate and how we interact with the universe around us. Intel Nervana deep learning technology provides open frameworks and open research to make it simple to deliver AI solutions while cutting edge research continues to advance the industry forward.  The FDL Lunar Water and Volatiles team, consisting of data scientists, software developers and planetary scientists, will be employing Intel Nervana technologies as part of their program. In addition, we’ve brought in Intel Principal Engineers and Data Scientists to accelerate development of the software solution and take AI (literally) to the moon.

We’d like to give a special shout out to our partner in the Lunar Water and Volatiles team, Space Resources Luxembourg. The country of Luxembourg has taken a leadership position in the utilization of space resources.

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

The project’s goal is to help future missions to locate water and volatile resources such as hydrogen, carbon dioxide, nitrogen and methane. These resources can be used to produce an air supply for astronauts, rocket fuel, and other essential materials. Locating and extracting volatile resources on the moon is a vital step towards future human space exploration to Mars and beyond. The team is answering two critical questions:

  1. Where are we likely to find water and volatiles?
Team Lunar Resources learned that there’s a wealth of data about the moon. We also have 50 years of optical imagery from the Apollo program and recent missions such as NASA’s Lunar Reconnaissance Orbiter (LRO). Hyperspectral data from India’s Chandraayan mission has indicated the presence of water in the permanently shadowed regions of polar craters. This supports the earlier NASA Lunar Crater Observation and Sensing Satellite (LCROSS) mission, where water was detected in the ejecta plume of a payload impact (the upper stage of LCROSS Centaur was purposely impacted into a crater). The cold and shadowed regions of the north and south poles are of particular interest for ’lunar prospectors’ as this is where important deposits of volatiles such as water ice are most likely to be found. Unfortunately, mapping the poles is a difficult task due to complex geometries thus making it hard to pinpoint where resources may be located.
  2. How do we reach these resources?
Even if we can find them, we still need a way to plan missions. The team has identified a critical knowledge gap in the management and quality control of the vast and growing quantity of lunar data.  Current maps are inadequate. In order to successfully conduct future landing missions, maps of the polar region must be improved.  After decades of research, the automated identification of lunar hazards such as craters, boulders and cliffs remains as one of the most important unsolved problems. This is required properly to inform planetary research and exploration missions with accurate and precise maps.

Our researchers will utilize deep learning methods to further the field of extracting features in the difficult polar regions using Intel Nervana technologies.  A showcase will demonstrate how deep learning can help to increase the quality and reliability of mapping data suitable to simulate and plan traverses (paths) for future lunar rovers.  Future commercial space missions to explore the lunar may depend on reliable information on where water and other volatiles may be found and the quantities that can be expected.

“Space data is often massive; multi-dimensional and dynamic. What’s more the kind of experimentation required to discover something new requires the ability to rapidly iterate on this data, so we can learn quickly and adapt as needed without having to wait too long. It’s for this reason that we reached out to our private sector partners who have the kit to allow FDL researchers to innovate in a highly accelerated time window.” – James Parr, FDL Director

What’s Next?

FDL 2017 kicked off in June and we’ve seen rapid progress. The teams tended to focus first on collecting existing resources related to their focus areas and mining them for key insights. The teams are treated to daily guest lecturers in their fields as well as visits to nearby space industry events (did we mention that they are working at the NASA Ames Research campus?). FDL has also been introducing the researchers to design thinking and lean startup principles, which are common in the tech industry, but new to them. Customer empathy and interviews are guiding the research implementations, making them more usable by the NASA mission planners and industry partners who will apply them. We have seen them iterate through a number of initial concepts, taking into account even late-breaking results, and ultimately solving actual problems for mission planners. We’re at the mid-point of the program now, and software development is underway. Be sure to come back for an update on the program next month!

A special moment for me, where I shared Intel’s vision with a room full of researchers, NASA  [Shashi] (Photo Credit: James Parr)

 

This blog was written by Shashi Jain and Katie Fritsch.

About Shashi Jain:

Shashi Jain is an Innovation Manager in the Software and Services Group at Intel, focused on pathfinding and corporate innovation in IoT, Machine Learning and Virtual Reality. He has over 20 years of experience in business development, entrepreneurship, engineering integration, rapid prototyping, and community building. In his spare time, he teaches innovation and entrepreneurship to high school students through the TYE Oregon and runs the largest 3D Printing Meetup on the west coast. Shashi has a Masters in Electrical Engineering from University of Illinois Urbana-Champaign and MBA from Babson College.

About Frontier Development Laboratory (FDL):

FDL is an applied artificial intelligence research accelerator and public / private partnership between NASA Ames Research Center and the SETI Institute. The program tackles knowledge gaps in space science by pairing machine learning expertise with astronomy and planetary science expertise at the PhD level.  Interdisciplinary teams address tightly defined problems and the format encourages rapid iteration and prototyping to create outputs with meaningful application to the space program.

About Intel Nervana:

Intel Nervana, leveraging Intel’s world leading position in silicon innovation and proven history in creating the compute standards that power our world, is transforming Artificial Intelligence (AI). Harnessing silicon designed specifically for AI, end-to-end solutions that broadly span from the data center to the edge, and tools that enable customers to quickly deploy and scale up, Intel Nervana is inside AI and leading the next evolution of compute.

 

Author Bio Image

Shashi Jain

Innovation Manager, Software and Services Group

Author Bio Image

Katie Fritsch

Sr. Ecosystem and Partnership Manager, Artificial Intelligence Products Group

Related Blog Posts

Machine Learning

A Summer of Space Exploration with Intel and NASA

This summer, Intel has been collaborating with the NASA Frontier Development Lab (FDL) , an AI R&D accelerator targeting knowledge gaps useful to the space program. The NASA FDL, hosted at the SETI Institute, was established to apply AI to five specific challenges in areas relevant to the space program: Planetary Defense (defending the Earth…

Read more

#NASA