Frontier Development Lab (FDL) – an artificial intelligence research accelerator for space science – has kicked off its 2020 program on a virtual platform with researchers and faculty from across the globe.
The Luxembourg Space Agency partners with FDL for the 4th consecutive year.
The teams, comprised of early-career PhDs in AI and interdisciplinary science domains, are supported by subject matter experts from NASA (including NASA Headquarters, NASA Ames Research Center and NASA Marshall Space Flight Center) and the SETI Institute together with FDL Partners: Google Cloud, Mayo Clinic, Lockheed Martin, MIT, USGS, IBM, Intel AI, Luxembourg Space Agency, NVIDIA, Planet and Augustus Intelligence. Along with expertise, FDL partner organisations support advanced AI research by providing funding, hardware, AI/ML algorithms, datasets, software and cloud-based super-compute resources.
“We celebrate the 5th anniversary of FDL with new partners, including the Mayo Clinic, MIT, USGS and Planet Labs – and new challenges as we enter the uncharted waters of executing FDL in a virtual environment,” said Bill Diamond, President and CEO of the SETI Institute. “We are matching these extraordinary times with extraordinary researchers, faculty, partners and research topics, and once again, we anticipate extraordinary results.”
Marc Serres, CEO of the Luxembourg Space Agency, said: “The Luxembourg Space Agency is excited to partner for the 4th consecutive year with the NASA Frontier Development Lab and organise a challenge on space resources. We are very curious to discover the high resolution images of the Permanently Shadowed Regions produced by the Moon for Good team in the challenge “Where is the ice? - AI-driven denoising of high-resolution PSR NAC imagery”, which can contribute to even better mission planning and target selection of future lunar surface missions. Acknowledging the positive energy that is needed to organise such a challenge under the current circumstances, we fully support the FDL 2020 edition in a virtual environment.”
INTERVIEW: Luxembourg, Guy Schumann - FDL mentor
FDL has demonstrated the potential of applied AI to deliver game-changing results to the space program, supported by a consortium of motivated partners. This approach has proven critical in unlocking meaningful progress in the complex and often systemic nature of AI problems.
Over the last five years, FDL has shown the potential for both scientific discovery and engineering, from predicting GPS scintillation to co-operative robots on the Moon. Additionally, the work informs better, earlier decisions, building maps, super-charging predictions and planning strategies within complex systems.
FDL’s goal is to apply these impressive new capabilities — many of which are emerging in the commercial sector — to problems important to space exploration and humanity.
The work produced by FDL researchers over the past four years has resulted in:
- 26 machine learning workflows
- 10 enhanced data products
- A #1 paper on arXiv in space physics
- A best paper at NeurIPS climate change and AI workshop
- Participation in 30+ scientific and AI conferences
- 12 papers presented at NeurIPS and 6 at AGU in 2019
- More than 60 blogs and articles in major media outlets
Teamwork is core to the FDL process. This year's program is comprised of a diverse group of over 100 researchers located around the world, representing important diversity in nationality, race and gender and more to bring their unique perspectives to this intense and rewarding work of developing cutting edge AI applications for space exploration and humankind.
About the NASA Frontier Development Lab (FDL)
Hosted by the SETI Institute, NASA FDL is an applied artificial intelligence research accelerator developed in partnership with NASA's Ames Research Center. Founded in 2016, the NASA FDL aims to apply AI technologies to challenges in space exploration by pairing machine learning expertise with space science and exploration researchers from academia and industry. These interdisciplinary teams address tightly defined problems, and the format encourages rapid iteration and prototyping to create outputs with meaningful application to the space program and humanity.