SCAN SUPPORTING AI SPACE RESEARCH AT FDL EUROPE 2023
Scan AI’s solutions and data science expertise leveraged to research problems important for space exploration and humanity
Scan Computers International, via its Scan AI specialist business unit that provides end-to-end solutions for deep learning and artificial intelligence, announced today that it is supporting this year’s AI space research sprint by Frontier Development Lab (FDL) Europe.
FDL Europe’s mission is to demonstrate the potential of AI for planetary stewardship, pushing the boundaries of Earth observation from world-first demonstrations of machine learning in orbit for mapping floods and landscape changes to predicting global-scale rainfall and understanding the drivers of extreme wildfires.
‘It is a privilege to be supporting FDL Europe research for the fourth year running and to be part of this great scientific initiative to tackle some of the biggest challenges facing humanity today’ stated Elan Raja, CEO of Scan, continuing ‘Scan AI has benefitted immensely from its involvement with FDL Europe in shaping some of its product offerings. The whole Scan AI team is excited to support such philanthropic space and climate projects and see the outcomes from this year’s research sprints’.
This year Scan will be providing both data science expertise and AI infrastructure to accelerate the research efforts of the FDL Europe teams. Our bespoke AI cluster comprises NVIDIA DGX systems and Omniverse Enterprise workstations connected to PEAK:AIO data servers, provisioned virtually from our Scan Cloud vGPU platform.
This summer, in partnership with ESA, Trillium Technologies selected three challenge teams to tackle cutting-edge research during this year’s 8-week research sprint: 1. Heliophysics Onboard, 2. Generalisable SSL SAR and 3. Foundation Model Adaptors for Disasters.
The Heliophysics Onboard challenge aims to use machine learning, onboard space observatories and on Earth, to build a near-real-time warning system for potentially dangerous solar weather events and to enhance our understanding of the solar system. The proposed AI pipeline will significantly shrink the required communications bandwidth between satellites and shorten alert latency, helping to safeguard astronauts and infrastructure.
The Generalisable SSL SAR challenge builds on last year’s DeepSlide challenge which used synthetic aperture radar (SAR) to detect landslides in four preselected areas of Puerto Rico, Japan, Indonesia and New Zealand. This further development aims to create precursor foundational models for analysing SAR images to understand how to create robust machine learning models in new locations.
The Foundation Model Adaptors for Disasters challenge aims to use the power of large language models (LLMs) that have garnered so many headlines in the last few months. The project focus is to further improve LLMs to curate, extract, interpret and dispatch key insights on natural disaster data with the goal of delivering information to disaster management teams in a more timely way with a particular focus on flooding.
Follow Scan AI on social media for updates as the FDL teams unveil the results of the three challenges over the course of the summer and visit our repository of AI case studies showing previous years’ challenges.
Press Contact:
James Gorbold, Director of Content
[email protected]
+44(0)1204 474791
About Scan:
At Scan we are all about three things – technology, people, and how we bring them together and for over 30 years we’ve been driven by these three passions. We work with customers and partners to tackle the latest cutting-edge areas in IT and bridge the gap these challenges present.
Scan AI is a specialist business unit that provides end-to-end solutions for deep learning and AI and is the UK's premier accredited NVIDIA Elite Solution Provider.
AI has the potential to revolutionise and improve business practice. However, for your business or organisation to embrace AI technology effectively a complete strategy is needed. Our ecosystem of GPU-accelerated hardware and cloud solutions deliver everything you need to see AI projects through development, training and inferencing. As every AI project is different we’re on hand with trusted advice and a host of additional services delivered by our expert team.
Find out more about Scan AI
About FDL (Frontier Development Lab):
FDL applies AI technologies to space science, to push the frontiers of research and develop new tools to help solve some of the biggest challenges that humanity faces. These range from the effects of climate change to predicting space weather, from improving disaster response, to identifying meteorites that could hold the key to the history of our universe.
FDL Europe is a public-private partnership between the European Space Agency (ESA), the University of Oxford and Trillium Technologies, in partnership with private industry partners such as Scan AI, NVIDIA and Google Cloud, which provide expertise and the computing resources necessary for rapid experimentation and iteration in data intensive research areas. FDL USA is a public-private partnership between NASA, the SETI Institute, Trillium Technologies and leaders in commercial AI, space exploration and earth science.
Find out more FDL Europe’s ground breaking research.