For the third time since 2016 machine learning experts and space scientists will spend the summer in Silicon Valley to work on some of NASA’s most important present day challenges. The 8 week long program - NASA frontier development lab (FDL) - at the SETI institute will face problems ranging from predicting solar storms to finding life in space. Explainable Daniel Angerhausen is an FDL science committee member and has been selected to facilitate this year’s lab and help mentor the Exoplanet and Astrobiology challenges. For the last week of the program Explainables Sally Blumenthal, Amran Salleh and Jesse John will join him to prepare the group for their final "TED-style" presentations.
The NASA FDL 2018 team during Kick-Off week at NVidia's Endevour campus.
On its webpage NASA frontier development lab describes itself as
“an artificial intelligence research accelerator established to apply AI technologies to challenges in space science & exploration for the benefit of all humankind. “
The program is a unique public / private partnership between NASA and commercial partners, such as Nvidia, Intel, IBM, Google and Lockheed Martin. These partners contribute the computational expertise and resources that are crucial for fast turnarounds in very data intensive areas. This year, FDL also has a challenge team in Europe supported by partners like Space Resources, Luxembourg and the European Space Agency, ESA.
“FDL’s special sauce is its interdisciplinary research teams - which are composed of two subject specialists from the space sciences and two specialists from the data sciences. The teams are handpicked by world class mentors and work together in an intensive sprint over the Summer.”, continues the description.
The explainables already worked with NASA FDL last year: they he prepared the teams for their final “TED style” presentations at the Intel Auditorium (see last years blog post). This year Explainables founder Daniel Angerhausen was selected to also serve as facilitator and mentor for the exoplanet and astrobiology challenges.
“It will be hard to miss most of the summer and the Aare swimming in Bern – but hey… if NASA asks to help humanity you can’t really say ‘No’” jokes Daniel.
“FDL last year was already a great experience and I am very much looking forward to seeing challenges from my background – Exoplanets and Astrobiology – this year. It is very inspiring to work on the interface of industry and academia with these big Silicon Valley companies and the world’s leading experts in Artificial Intelligence, but what excites me the most is interacting with the pool of young bright minds from all over the world.”
This year’s Astrobiology challenges are about understanding what is universally possible for life and how we can get from biohints to confirmed evidence of life on exoplanets within given environmental substrates. In the Exoplanets part the goals are to use AI/ML to increase the efficacy and yield of exoplanets detection from TESS and to codify the process of AI derived exoplanet discovery and classification to help TESS better discern rocky planets.
Other challenges come from the areas of SPACE RESOURCES (e.g. Modelling the space Resources economy in its first 50 years), SPACE WEATHER (e.g. Predicting solar spectral irradiance from observations), SPACE TRAFFIC MANAGEMENT / ORBITAL DEBRIS (e.g. Predictive model for atmospheric density and spacecraft drag), and EARTH OBSERVATION (towards a ‘Mission Control’ for Planet Earth).
For more details check out: NASA frontier development.
