Skip to main content
conference lunch move company map contacts lindholmen lindholmen 2 travel info

Φ-lab @Sweden on edge learning in space

The European Space Agency (ESA) and AI Sweden are establishing a space research lab, Φ-lab @Sweden, to accelerate the use of AI in new applications for Earth observation. The lab is the spearhead of a major European collaboration on space innovation and will be inaugurated in the spring.

screenshot_2022-08-05_at_15.14.01_0.png

Geraldine Naja, Director of Commercialisation, Industry and Procurement (D/CIP), at the launch of the Edge Learning Consortium, Φ-lab @Sweden, and Cyber Security Lab.

The Φ-lab @Sweden gathers industry, investors, and researchers to strengthen the European space research and space industry sector in Earth observation and AI. The work complements the efforts at the Swedish Space Data Lab, which includes the Swedish National Space Agency, Luleå University of Technology, RISE and AI Sweden.

Why edge learning in space?

Traditionally, good AI models are trained by transferring as much data as possible from sensors to a central storage and compute location. In general, the more data, the better the quality of the resulting model. However, as the number of satellites increases, the available bandwidth to transfer data will become less and less. Thus, data transfer cost is expected to rise, it will take longer to transfer enough data to earth to train models, and the models will be less accurate if less data is transferred. 

Edge learning as a technology has the potential to revolutionize the development of AI for space applications by solving these challenges. This would save both money and resources since there is no need for sending all the large amounts of data to the Earth. Possible use cases for AI and satellite data include identifying flooding, illegal fishing, deforestation, algal blooms, or environmental disasters.

Aggregation and Swarm Topology

Aggregation Topology

  1. Create initial model on earth
  2. Transmit current model
  3. Training on satellites using local data
  4. Transmit local model parameters to earth
  5. Aggregate parameters to create updated model
  6. Repeat cycle from step 2

Swarm Topology

  1. Satellites register on a network
  2. Satellites receive an initial model
  3. Satellites train model on local data
  4. Satellites share and merge models
  5. Repeat steps 3 and 4 until satisfied.

How to get involved

Get in touch with Ebba Josefson Lindqvist if you want to learn more about the Φ-lab @Sweden and how you could get involved.

What is a Φ-lab?

As part of the Boost Green and Digital Commercialisation priority of its Agenda 2025 strategy, and building on the experience of the existing Φ-lab @ESRIN with its focus on delivering transformative innovations aimed at market adoption, ESA is creating a dynamic network of Φ-labs across ESA and Europe. The Φ-lab @Sweden is the first to be established within this larger network.

More about the ESA Φ-lab

More about ESRIN

Contact

Project Manager, LLM

Ebba Josefson Lindqvist

Mats Nordlund
Head of Data Factory, PhD Eng

Mats Nordlund