The Internet of Space

At EDPS back in May (yes, I know I'm behind), Andrew Filo talked about Building the Internet of Space (IOS). He is an independent consultant but works with places like NASA Ames and was a consultant on the 3D printer in the ISS. What he was really talking about was building what he called femto-satellites (or sometimes chipsats) and putting them into space in comparatively large numbers. This project originated with the Cornell University Space Systems Design Studio.

Femto-Satellites

So what does a femto-satellite look like?

That is probably smaller than what you were thinking of. It is to a normal satellite rather like Raspberry Pi or Arduino is to a supercomputer. The plan is to switch to an Intel Curie since it packs a lot of functionality into a tiny space.

So why might you want to put a bunch of femto satellites into space? Andrew gave several reasons:

• Communications
• Broad coverage of earth
• Sense and monitor objects in space
• Monitor localized space weather conditions (such as solar flares)
• Monitor localized cosmic radiation
• Provide navigational reference
• Generate secure communication protocols

The Kicksat

So how do you get the femto-satellites up there? Through a sort of frame called a kicksat that holds 128 of the femto-satellites and an Intel development board that controls the deployment. The frame looks like this:

The first kicksat was launched in April 18, 2014, amazingly funded through Kickstarter, on an ISS commercial re-supply launch. The chipsats were planned to orbit for two to six weeks before they burn up re-entering the earth's atmosphere.

But after a couple of weeks, the microcontroller managing the master clock was found to have reset due to a technical problem, an effect of space radiation. This reset added two weeks to the deployment schedule for the femto-satellites. Kicksat's battery needed to get enough power to allow deployment of the sprites before KickSat itself began atmospheric reentry. But it was not to be, and on May 14, 2014, KickSat reentered the atmosphere and burned up; all the femto-satellites were lost.

Kicksat-2

Kicksat-2 will be launched sometime this year under the auspices of Cornell University. NASA announced in 2015 that it had selected KickSat-2 for launch as part of its CubeSat Launch Initiative. As the NASA press release says:

KickSat-2 is a CubeSat technology demonstration mission designed to demonstrate the deployment and operation of prototype Sprite “ChipSats” (femtosatellites). The Sprite is a tiny spacecraft that includes power, sensor and communication systems on a printed circuit board measuring 3.5 by 3.5 centimeters with a thickness of a few millimeters and a mass of a few grams. ChipSats could enable new kinds of science and exploration missions, as well as dramatically lower the cost of access to space.

Since it is December, there is not much more "sometime" left this year, but Googling around I couldn't find any evidence that it has yet launched.

At EDPS, Andrew had several of the femto-satellites that we passed around.

Hopefully the second deployment will be more successful than the first. It certainly seems like something interesting to keep an eye on. Swarms of $100 satellites sound pretty attractive compared to the multi-million dollar cost of a conventional satellite, although obviously more restricted. Plus you need to relaunch a new swarm every month or two for continuous coverage.

Take a look at the NASA page describing the Kicksat-2 project. It was last updated in November.

Next: RISC-V 5th Workshop Highlights

Previous: Protium: FPGAs the Cadence Way