DataRobot header v.1.0

A Space-Focused Start-up Takes Aim at Space Junk

April 12, 2021
by
· 4 min read

The past year has given us plenty to worry about, but I’ll add one more thing to the list: space junk.

That’s because the skies far above us harbor countless bits of human-made objects. Although many are tiny, because they travel at 20,000 miles per hour, a collision with one of the more than 8,000 earth-orbiting satellites up there could be catastrophic.

I recently had the opportunity to speak with two entrepreneurs who are working to make space a safer place for satellites and, by extension, the things they do for us, from tracking hurricanes to transmitting internet data to giving us directions while we’re driving.

Siamak Hesar and Araz Feyzi are the co-founders of Kayhan Space, a start-up that helps manage the rapidly growing number of earth-orbiting satellites. Over the next few years, Amazon plans to launch 3,236 satellites in its Kuiper network. SpaceX plans to launch up to 30,000 satellites. In the words of TechCrunch, “That’s a lot of metal flying around.”

Hesar is an expert in flight dynamics and space flight mechanics, areas he studied while earning his PhD at the University of Colorado Boulder. Feyzi’s specialty is product development and cloud computing. Friends since high school and college roommates, the two were pursuing different career paths when they got to talking about Hesar’s frustration with the state of software in the space industry. “I’ve seen cases where people write software on a CD and ship it to others,” he said to me. “In this age of cloud computing and high-performance computing, that doesn’t make sense.”

Feyzi was intrigued. “That sounded like a great business possibility,” he said. “Since technology is my hammer, I offered to get together with him to see if we could offer a product that would help solve problems.”

Although we’re used to thinking of it as infinite, Hesar says, “Space is a very large area, but operationally the layers of satellite orbits are really quite narrow (between 100-1,200 miles).  When you put 30,000 satellites in that narrow range, it becomes congested. That’s a real concern these days.”

Avoiding the Kessler Syndrome

In time, we might face the Kessler Syndrome,  a chain reaction the former NASA scientist, Don J. Kessler, warned of in a 1978 paper. When the density of objects in low-earth orbit reaches a certain point, the collisions of satellites create more debris, creating more collisions, and so on. As Kessler said, “That could make space unusable for hundreds of years.”

To date there has been only one serious in-space collision. In 2009, an Iridium Communications satellite and a Soviet-era satellite collided 500 miles above Siberia, creating thousands of pieces of debris. There have been several close calls, however. In October 2020, a dead Soviet-era satellite narrowly missed a Chinese rocket stage. In 2020 alone, the International Space Station had to maneuver three times to avoid collisions with another object.

In fact, in any given week, a satellite operator might deal with up to 100 collision warnings.

Kayhan Space wants to use its software in conjunction with existing satellite-tracking networks to simplify collision avoidance to the point it becomes a Slack workstream. 

For Hesar and Feyzi, improving space management requires:

  • Improving the current network of satellite-tracking stations. “The more [tracking] coverage you have in space, the better it is in terms of tracking and advance notice,” says Hesar.
  • Improving the modeling of space. Space is thought of as a vacuum, but in low-earth orbit, satellites encounter a fair amount of atmospheric drag. And even at very high elevations, solar radiation creates just enough force to alter the flight of satellites.
  • Improving responses to imminent collisions. In September 2019, a SpaceX satellite nearly smacked into a European Space Agency (ESA) craft. Although a U.S. Air Force tracking station alerted ESA to the danger, a flaw in SpaceX’s alert system meant no one there was aware of the problem. As Hesar deadpans, “Having humans in the loop creates more chances of error.” 

Automating Collision Avoidance

The goal of Kayhan Space is to automate that process. Feyzi says, “Most satellite operators want more automation and more autonomy. That’s something that’s in our wheelhouse so we’ve prioritized it.” Achieving that goal will require sophisticated AI modeling and machine learning.

In time, in addition to autonomous satellite control, Feyzi says he and Hesar envision something like a giant dashboard that contains real-time information about the locations of debris particles and satellites and calculates the risk of collisions. They might even investigate the use of virtual reality.

Still in start-up mode, Kayhan Space recently received $600,000 in seed funding and will gradually expand its current workforce of eleven employees. “It’s a fun team,” says Feyzi. “We have people all the way from New York state to Oregon.”

And of course, they’re working in a field they find tremendously exciting. “A lot of kids say, ‘I want to be an astronaut,’ and I was one of those kids,” Hesar said. “That kind of passion stuck with me. I hope this company leaves a legacy that makes a difference in our lives.”

To learn more about AI, check out Datarobot.com/podcast or https://datarobot.buzzsprout.com/. You can also listen everywhere you already enjoy podcasts, including Apple, Spotify, Stitcher, and Google.

Podcast
Siamak Hesar and Araz Feyzi: Tracking space hammers with AI
Watch Now

About the author
DataRobot

Value-Driven AI

DataRobot is the leader in Value-Driven AI – a unique and collaborative approach to AI that combines our open AI platform, deep AI expertise and broad use-case implementation to improve how customers run, grow and optimize their business. The DataRobot AI Platform is the only complete AI lifecycle platform that interoperates with your existing investments in data, applications and business processes, and can be deployed on-prem or in any cloud environment. DataRobot and our partners have a decade of world-class AI expertise collaborating with AI teams (data scientists, business and IT), removing common blockers and developing best practices to successfully navigate projects that result in faster time to value, increased revenue and reduced costs. DataRobot customers include 40% of the Fortune 50, 8 of top 10 US banks, 7 of the top 10 pharmaceutical companies, 7 of the top 10 telcos, 5 of top 10 global manufacturers.

Meet DataRobot
  • Listen to the blog
     
  • Share this post
    Subscribe to DataRobot Blog
    Newsletter Subscription
    Subscribe to our Blog