FDL 2019 is your chance to be part of something ground-breaking this year.
NASA fdl 2019 challenge areas
The FDL yearly cycle starts with challenge definition. Early in the year, we bring together some of the brightest and best minds we can find, from space science, AI and technology, and on/off-Earth applications to explore our challenge areas. During the course of our day-long Big Think events in Europe and the US, we aim to identify some broad challenges, which the FDL research teams could tackle in the summer.
Through a process of iteration with a PI (principal investigator) leading each challenge, we refine and narrow those challenge areas until we have identified one, or several, tightly articulated questions to resolve.
FDL challenges must represent a clear and present scientific problem, for which there is available data, that could be significantly advanced by AI tools and techniques. It is these challenges that the research teams further narrow in the opening weeks of the FDL research sprint to refine their own particular concept approach. The broad challenge areas we start the year with move from provisional to confirmed as we understand how, and when, they meet these criteria. Our provisional challenge areas for 2019 are:
Living with Our Star
How can AI improve our ability to predict solar activity - especially energetic solar phenomena?
FDL has already demonstrated the game-changing potential for AI techniques to provide predictive capacity in the field of heliophysics. Outputs of previous research sprints have shown that AI can be used to predict C-class flaring events, the Kp Index, GPS scintillation and UV irradiance.
This raises the tantalizing possibility of using AI to provide a full-spectrum forecast of solar activity many hours in advance. Such a forecast would be of enormous value to future deep-space exploration, where solar activity remains a high-risk, but hard to predict variable. The potential benefit of improved solar weather forecasts is of great value on Earth too, since crucial infrastructure such as communications and power networks can be impacted by solar flares. Improved prediction and characterization of space weather is also central to the pursuits of optimal space situational awareness.
Need being addressed: Safeguarding of our ground- and space-based infrastructure, through better prediction of solar events, allowing time to take evasive or remedial action.
Technical angle: Deep learning models which leverage large amounts of multi-variant data could be used to enhance predictive efficacy.
The Moon for Good
How might AI support the goal of establishing a permanent presence on the Moon?
NASA has ambitions to establish a permanent human presence on the Moon in partnership with commercial operators. FDL has a growing toolbox of AI tools for lunar exploration which includes crater identification, lunar navigational aids and methods for cooperative robotics for polar prospecting on the Moon. This year FDL will build on our existing work, and begin to extend our portfolio to include human-robotic collaboration, swarm robotics and evolve resource mapping techniques.
Need being addressed: How can the needs of the private sector to establish a presence on the Moon be pushed forward using AI capabilities? Can AI enable other lunar missions?
Technical angle: Autonomy for exploration and building off previous FDL work in robotics and lunar mapping.
Are We Alone?
Can AI help answer the question of whether we are alone in the universe?
FDL and the SETI Institute are already exploring the application of AI to the search for extraterrestrial signals (sometimes known as ‘technosignatures’). The large data sets and improved computing power now available means that new opportunities for large-scale analysis are now emerging. One example of this is from Francois Luus of IBM Research, South Africa, who has developed an LSTM neural net model to analyze data collected by the SETI Institute's Allen Telescope Array. This could form the starting point for more ambitious and scalable implementations during FDL 2019
Need being addressed: First large-scale application of emerging AI tools (such as single-shot learning) to the search for technosignatures.
Technical angle: Application of anomaly detection and self-learning methodologies.
Mission Control for Earth: Real-Time Data/Virtual Instruments
How might we utilise AI and Earth observation data to support improved decision making to protect the planet?
FDL aims to create useful tools that can help solve big challenges on Earth, and we like to envision a “Mission Control Centre” for Earth, that would give real-time insights on the health of the planet and allow us to prepare for the impact of events such as storms, earthquakes, forest fires etc more efficiently.
If we could derive insight into our environment in real-time and predict future scenarios we would be empowered to take action and provide early-warnings that could save lives. The data is already becoming available, but to realize this vision, we still need to find how to extract value from it quickly. The ability to pull out insights in real-time is still limited, but AI promises to be a powerful tool for sense-making, data fusion and making reliable predictions.
Need being addressed: Extracting useful insights from data in real-time, to enable early and improved interventions, for example identifying the first signs of wildfires before they have chance to spread, or developing predictions that could enable illegal poaching to be interrupted.
Technical angle: New era of SmallSats and satellite constellations are providing high frequency coverage opening up a new era of applications that require rapid insight. Applying AI techniques to enable value to be extracted from data-sets in real-time.
How can AI support medical care in space?
NASA plans to recommence human missions beyond low-Earth orbit (LEO) and carry out longduration missions in deep space. The distances involved mean that crews will need to manage manage their health autonomously, since communication with the ground may be subject to significant delays. This requires a paradigm shift in how medical care is provided to astronauts, moving from Earth-reliant (telemedicine and evacuation) protocols to Earth-independent strategies (long-term medical autonomy).
Need being addressed: As we continue to expand human exploration for longer durations, and further out into the solar system, we will require more advanced solutions for monitoring and maintaining human health.
Technical angle: This challenge area explores the potential to train an AI model to produce more natural vascular systems based on a large image database of real vascular networks or explore the ability of Astroskin data to train a generative AI model to produce additional synthetic biosensor data.
You can read about the FDL Europe 2019 challenges by following this link.