A graphing calculator has more processing power than most satellites. To pursue the next generation of space missions, operators will need better and faster compute capabilities in orbit. In this episode, Edward Ge, co-founder and CEO of Aethero, discussed what it will take to build this type of compute, how it will benefit the industry, and what challenges still lie ahead.
Listen to the full podcast episode, or read our top four takeaways below.
Takeaway 1: Modern satellites need to be able to handle more advanced computing.
“The same processors that were powering your computer back before the dot com era are the same processors that are powering NASA’s Mars rovers,” Ge said. While many in the industry have begun discussing how to incorporate artificial intelligence into satellite systems, Ge emphasized that the first step towards more advanced space missions is building more advanced compute capabilities into satellites.
“The way we run our satellites hasn’t changed much since the 1980s,” Ge said, which includes extremely limited downlink time. “You only [have] 15, 20 minutes that you can actually get that data down from the satellite, look at the data, and then send commands back up,” he said. But if you instead have enough compute power to compress and process the data, and then make an autonomous decision about what to do next, “that essentially enables a satellite to react dynamically and in real-time to changing events.”
“That saves you a lot of time, that saves you a lot of effort, and that could prevent a multi-million dollar or multi-billion dollar asset from being damaged or destroyed,” Ge said.
Takeaway 2: Using COTS components supports faster innovation and market adoption.
Developing commercial off-the-shelf products will be a key part of building better computing systems on-orbit. “The movement towards COTS components is something that’s really been pioneered by the new wave of space-age startups that have appeared since the 2010s,” Ge said.
“COTS components are surprisingly resilient in space,” Ge said, and specifically noted the benefit of using off-the-shelf technology for short-term missions. “I do think there is a trend right now towards more satellites in space that have disposable lifetimes,” he said. This movement towards shorter mission lifetimes is driven by the drop in launch cost created by SpaceX’s Falcon 9 and other commercial launch providers.
Takeaway 3: On-board processing will enable technologies like Golden Dome and ISAM.
“A lot of the technologies needed for advanced missile defense and space interceptors cannot happen without onboard compute, and cannot happen without onboard autonomy,” and will be crucial for a program like Golden Dome. That type of heavy-duty compute will also drive on-orbit ISAM capabilities.
“A lot of advanced use cases like ISAM, autonomous RPO, and even defense use cases all depend very heavily on your satellite being able to react in real time,” Ge said. Ground-based processing simply cannot handle the speed and amount of data processing required for these activities. Because satellites only have contact with the ground for 15 to 20 minutes at a time, “there’s a massive window of every orbit where you can’t get the data down, and if you can’t get the data down, there’s nothing to run the compute on,” he said.
Currently, “satellites are limited by transmit power, by transmit frequency and duty cycles,” Ge said, which is what drives the need for on-board compute power.
Takeaway 4: The regulatory environment lags behind the technology.
While the technology for on-board computing continues to advance, the regulatory environment has only a sparse framework. “It’s one of the fields where it’s so far ahead that there is not much of a regulatory environment,” Ge said.
Crucially, the FCC does track information related to orbital debris, as congestion has become a more pressing issue for satellite operators across the world. “No one wants Kessler syndrome to happen, and to knock us out of low Earth orbit for the next decade,” Ge said.
But in terms of actual space activities, there is very little regulatory guidance. “There isn’t that much yet in terms of regulatory overhead,” he said. “And that’s good. It gives us room to innovate, it gives us room to experiment, and it let’s us figure out what works and what doesn’t work before very single satellite in orbit is AI-powered.”
For more on computer processors, defense applications and lunar missions, listen to the full podcast episode.
Explore More:
That Computes! Edge Computing Reshapes Space Possibilities
Podcast: Old Versus New, Plug and Play and Responsive Satellites
Data Centers in Space – Why Data Processing is Moving from the Ground to On-Orbit
