MENU


The company, along with the European Space Agency (ESA) and computer vision and AI startup Ubotica, announced details of the PhiSat-1, an experimental CubeSat satellite that was ejected from a rocket’s dispenser on September 2 along with 45 other similarly small satellites. PhiSat-1 contains a new hyperspectral-thermal camera and onboard AI processing – thanks to an Intel Movidius Myriad 2 Vision Processing Unit (VPU), says the company, the same chip inside many smart cameras and even a $99 selfie drone.

PhiSat-1 is one of a pair of satellites on a mission to monitor polar ice and soil moisture, while also testing inter-satellite communication systems in order to create a future network of federated satellites. The first problem the Myriad 2 is helping to solve, say the organizations, is how to handle the large amount of data generated by high-fidelity cameras like the one on PhiSat-1.

“The capability that sensors have to produce data increases by a factor of 100 every generation,” says Gianluca Furano, data systems and onboard computing lead at the European Space Agency, which led the collaborative effort behind PhiSat-1. “While our capabilities to download data are increasing, but only by a factor of three, four, five per generation.”

At the same time, about two-thirds of the Earth’s surface is covered in clouds at any given time. That means, say the organizations, that a lot of useless images of clouds are typically captured, saved, sent over precious down-link bandwidth to Earth, saved again, reviewed by a scientist (or an algorithm) on a computer hours or days later — only to be deleted.

“And artificial intelligence at the edge came to rescue us, the cavalry in the Western movie,” says Furano.

The idea the team rallied around was to use onboard processing to identify and discard cloudy images – thus saving about 30% of bandwidth.

“Space is the ultimate edge,” says Aubrey Dunne, chief technology officer of Ubotica, which built and tested PhiSat-1’s AI technology. “The Myriad was absolutely designed from the ground up to have an impressive compute capability but in a very low power envelope, and that really suits space applications.”

The Myriad 2, however, was not intended for orbit. Spacecraft systems typically use very specialized radiation-hardened chips that, says Dunne, can be “up to two decades behind state-of-the-art commercial technology.” And AI has not been on the menu.

As a result, Ubotica performed “radiation characterization” on the Myriad chip, putting it through a series of tests to figure out how to handle any resulting errors or wear-and-tear.

“[ESA] had never tested a chip of this complexity for radiation,” says Furano. “We were doubtful we could test it properly … we had to write the handbook on how to perform a comprehensive test and characterization for this chip from scratch.”

The tests however, which included 36 straight hours of radiation-beam blasting at CERN in late 2018, turned out well. The Myriad 2 passed in off-the-shelf form, with no modifications needed.

Another challenge, say the organizations, was finding training data for AI chip, which was new.

“We didn’t have any data,” says Furano. “We had to train our application on synthetic data extracted from existing missions.”

The system and software integration and testing, with involvement of a half-dozen different organizations across Europe, took four months to complete. Once launched, as an initial verification, the satellite saved all images and recorded its AI cloud detection decision for each, so the team on the ground could verify that its implanted brain was behaving as expected.

Late last month, ESA announced the joint team was “happy to reveal the first-ever hardware-accelerated AI inference of Earth observation images on an in-orbit satellite.”

By only sending useful pixels, the satellite will now “improve bandwidth utilization and significantly reduce aggregated downlink costs,” as well as save scientists’ time on the ground. Looking forward, say the organizations, the usages for low-cost, AI-enhanced “teensy” satellites are innumerable – particularly when the ability to run multiple applications is added.

“Rather than having dedicated hardware in a satellite that does one thing,” says Jonathan Byrne, head of the Intel Movidius technology office, “it’s possible to switch networks in and out.”

For example, when flying over areas prone to wildfire, a satellite can spot fires and notify local responders in minutes rather than hours. When flying over oceans, which are typically ignored, a satellite can spot rogue ships or environmental accidents. Over forests and farms, a satellite can track soil moisture and the growth of crops; and over ice, it can track thickness and melting ponds to help monitor climate change.

Looking ahead, ESA and Ubotica are working together on PhiSat-2, which will carry another Myriad 2 into orbit. PhiSat-2, say the organizations, will be “capable of running AI apps that can be developed, easily installed, validated and operated on the spacecraft during their flight using a simple user interface.”

Intel
European Space Agency
Ubotica

Related articles:
Self-driving car AI tech to power space electronics
Mission-level autonomy coming to DARPA small satellite network
Smart satellite tests ‘new era’ of space-based computing

 


Share:

Linked Articles
Smart2.0
10s