NASA Frontier Development Lab (FDL), hosted at the SETI Institute, kicked off its 2017 program on Monday.

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 experts with astronomy and planetary science expertise. Interdisciplinary teams address tightly defined problems with meaningful application to the space program.

Participants are set up in five teams consisting of four members:

• Planetary Defense: Long Period Comets
  Provide more warning time for long-period comet impacts by applying deep learning to meteor shower observations
• Planetary Defense: Radar 3D Shape Modeling
  Develop a methodology to automate the backlog of neo radar imagery that requires shape modeling – and also improve the resolution of the result
• Space Resources: Lunar Water & Volatiles
  Determine the location and most promising access points for vital lunar H2O, in terms of cost effectiveness and engineering constraints
• Space Weather: Solar-Terrestrial Interactions
  Improve understanding of solar influence on Earth’s magnetosphere and atmosphere
• Space Weather: Solar Storm Prediction
  Discovering new relationships and agents to help predict major solar events

A sixth team comprised of undergraduate students will address AI and the space sciences by exploring the application of AI as a breakthrough capability for the space program, informed by the experience of FDL and its partner network.

The 2017 participants come from around the world including the United States, Belgium, Brazil, Colombia, the United Kingdom, Australia, Finland, India, and Luxembourg. Teams will be mentored by planetary scientists, data scientists, and private sector experts.

Core project partners of FDL include IBM, Intel, KX, and SpaceResources Luxembourg. Founding FDL partners are Autodesk and Nvidia. Research partners are Lockheed Martin, USC, and XPrize.

The format encourages rapid iteration and prototyping to create outputs with meaningful application, papers, and conference posters.

Teams will work through several phases, from concept definition to data prep and prototyping to concept demo, culminating in team presentations at the NASA AI Conference in August.