PhiSat – First Earth observation satellite with artificial intelligence soon to launch

The first Earth observation satellite with artificial intelligence is set to launch a few months from now, according to the European Space Agency (ESA) on September 12, 2019.

The idea was proposed by the Polytechnic University of Catalonia in Spain and won the Copernicus Masters competition in 2017. It was then named ɸ-Sat or PhiSat.

ESA has been working with several partners to improve the PhiSat to see the probability of artificial intelligence in space and take the mission on a further level.

Josef Aschbacher, ESA's Director of Earth Observation Programs, said that there is a huge interest in PhiSat. He expressed his gratitude to the agency's partners who helped develop the idea until it is ready to be launched soon.

The artificial intelligence technology will fly on one of the two CubeSats that complete the FSSCat Mission. It will be used to show how AI can upgrade the effectiveness of sending Earth observation data back to the planet.

The CubeSats, each with the size of a shoebox, will gather data that will be made accessible through the Copernicus Land and Marine Environment services using a cutting-edge dual microwave and hyperspectral optical instruments. They also carry a set of inter-satellite communication technology experiments.

The hyperspectral camera on one of the CubeSats will gather a massive number of images of Earth and to prevent downlinking the images back to Earth, the AI chip will filter them out so only the ones fit to use are sent back.

Aschbacher remarked, "We live in exciting times, the pace at which digital technology is developing coupled with the wealth of satellite information being delivered and, indeed, the growing demand for such data, means there are many opportunities to make a step change for the future of Earth observation."

He added that this would be made possible with PhiSat, Europe's first AI in space.

Marco Esposito from Remote Sensing, the company that spearheaded the upgrades of the AI algorithm, said that "while compact, the instrument– which covers the visible and near infrared with hyperspectral capability, enhanced with bands in the thermal infrared – is very powerful and will acquire terabytes of data that can be used to monitor vegetation changes and to assess water quality, for example."

However, Esposito explained that producing that number of information may trigger an error so the data have to be managed efficiently.

"With ɸ-Sat, we have effectively given the instrument its own brain, which processes the data onboard to detect clouds in the images. This not only ensures better quality data but makes the delivery much more efficient," he explained.

Massimiliano Pastena from ESA confirmed that the AI will be launched in the coming months, saying that the team is looking forward to it.

"I am very proud that we at ESA and our partners are putting Europe at the forefront of this new approach for Earth observation. And, as I announced at the opening of ɸ-week, our efforts are not stopping at ɸ-Sat-1– we are also soon going to release a new challenge to develop ɸ-Sat-2," Aschbacher said.

Featured image credit: European Space Agency