Skip to main content


It's hard enough to get busy families to stay connected, but what about a constellation of communicating satellites hurtling at high speed through orbit many miles apart? Gabor Karsai, professor of electrical engineering and computer science and associate director of ISIS, is leading a team creating just such a network in the sky called a "fractionated spacecraft," which is a cluster of independent small satellites that work together to perform coordinated missions.

The System F6 program is a cooperative effort of government, academia, industry, and nonprofit entities funded by DARPA with NASA acting as technical supervisor. The goal of the program is to demonstratethe feasibility and benefits of a fundamentally new satellite architecture. In this architecture, the functionality of a traditional "monolithic" spacecraft is delivered by a cluster of wirelessly interconnected modules capable of sharing their resources elsewhere in the cluster. The first flight test of a System F6-based fractionated spacecraft is set for 2015.

The original idea behind the System F6 program sprang from the notion that many small satellites can perform a single task better than a single satellite. That kernel of an idea expanded to the concept of networking many small satellites together to perform multiple tasks. A fractionated satellite architecture offers more flexibility and robustness than traditional design during mission operations, as well as during design and procurement. In particular, fractionation enhances space system adaptability and survivability, while shortening development timelines and reducing the barrier-to-entry for participants in the space industry.

"The big problem with single satellites is that if something goes wrong, you have a very expensive piece of space junk," Karsai said. "With the System F6 model, you have many smaller satellites, with every one doing part of the work. You have plenty of spares so that when something does go wrong—any one can fail and failures do occur—the network survives."

One of the crucial scientific and engineering challenges in creating fractionated satellites is the robust information system architecture. Of the eight groups competing for the software portion of the System F6 project, the Vanderbilt ISIS team won with a unique ability to integrate every piece of the puzzle to invent an entirely new and comprehensive software platform for fractionated spacecraft during the project's initial phase, according to Karsai.

After the initial design phase, ISIS was tapped to over-see the completion of the System F6 information architecture platform. Additional members of the distributed team building this system include Kestrel Institute from Palo Alto, California; Object Computing Inc., from St. Louis, Missouri; Remedy IT from the Netherlands; and Saffire Systems from Indianapolis, Indiana.

"This project is not only about designing and writing a new networked operating system," Karsai explained. "We have to create complex security safeguards that prevent different software programs from interfering with each other. We have to quickly handle faults that appear all the time in a complex system such as this. We have to manage the network that is subject to drastic degradations and data loss. All these problems must be solved and addressed together. You can't address them one by one in isolation."

ISIS has the critical mass to tackle such complex problems because it can quickly bring together a diverse team, he said. "We were able to put together a team that not only has experience with design, but also with building software and doing complex systems engineering. We could have taken a very narrow design view and just focused on the software. But we have to worry about how the software interacts with the spacecraft and how faults on the spacecraft will impact the software. That's not something a typical computer scientist does," Karsai said.

He obviously relishes the challenge and likens the current task to the early pioneering days of the Internet, which was likewise funded by the government, starting in the late 1960s and early 1970s. Those inventors could scarcely imagine the scope and impact of the Internet today.

DARPA is taking a similar approach to the System F6 program, envisioning these networked satellites as a Space Global Commons. Industry standards (such as the Internet Protocols) and open-source platform technologies (such as the Linux operating system and CORBA and DDS middle-ware) form the basis of the network, creating the building blocks for future space applications. In particular, the ACE, TAO, and CIAO open-source middleware (see page 30) used on System F6 have been guided by Douglas C. Schmidt, professor of computer science and senior research scientist at ISIS, and Aniruddha Gokhale, associate professor of computer science and senior research scientist at ISIS.

"Our goal is to create an open standard where any vendor can build a satellite that can fully participate in a fractionated satellite cluster," Karsai said. Similarly, the open-source information architecture platform allows inventors to develop all kinds of applications, creating myriad spin-off opportunities. The System F6 software can be used to implement commercial applications, as well.

The System F6 operating package under development by ISIS could one day be integrated into existing satellites, bringing them into a System F6 cluster without having to build new spacecraft, Karsai explained. Even the inexpensive CubeSat technology—often used to fly educational payloads—could be brought into a network.

"The System F6 system software must be flexible for use in a range of different topics and missions," Karsai said. Hence, the system puts into practice the full span of complex computer science problems and topics. Eventually, the goal is for the System F6 system to grow into a self-sustaining, self-motivating ecosystem populated with contractors, problem solvers, and those in need of solutions.

The spin-offs resulting from the System F6 project have the potential to perform many complex research projects by pointing sensors in any direction—toward space, the atmosphere, the oceans, earth's crust—taking constant measurements and sending back data to help scientists better understand how wheat grows or how to site an archeological dig, for example. As Karsai said, "It's hard to imagine all of the possibilities for growth. We're literally building a business incubator for R&D in the sky."

fractionating satellites