Project Suncatcher is Google's newest "moonshot" project, a groundbreaking project that aims to put Machine Learning (ML) machines on the ground and run them in space.

In this project, Google Solar-powered satellites to TPUs (Tensor Processing Units) combined with free-space optical links It is connected to technology.
The goal is to create a “space-based AI supercomputer” that can simultaneously run compute on the same scale as data centers on Earth. 

Project Suncatcher

Project Suncatcher's vision

AI technology is creating a new generation and achieving many scientific and economic achievements.
But on Earth, due to space, energy, and the environment, AI compute It's getting more difficult to expand. 

Project Suncatcher is a very nice way to solve this problem.
Space could provide solar power all day long and not require batteries, making it a more efficient computing environment. 

 

System Architecture – AI Satellite Clusters 

Project Suncatcher low Earth orbit It aims to build a network of solar-powered satellites in low-orbit space.
Each satellite contains Google's TPU chips, optical link network And will contact each one. 

Component  Function 
Solar Arrays  Permanent energy generation through solar 
TPU Modules  Optimized hardware to run Machine Learning 
Optical Links (DWDM)  High-speed, low-latency transport 
Cluster Formation  Grouping satellites just a few meters apart 

This approach results in a modular AI infrastructure that can be easily expanded and has redundancy (backup capability). 

 

Suncatcher Facing Engineering Challenges 

  1. Inter-Satellite Communication Speed
    The amount of bandwidth required to run Machine Learning workloads in a space like a data center is simply enormous.
    Suncatcher Dense Wavelength Division Multiplexing (DWDM) With technology 1.6 Tbps It has successfully passed data transmission tests and is fully practical with stable formation control technology. 
  2. Orbital Dynamics & Cluster Stability
    To control satellites from a distance of just a few meters Hill-Clohessy-Wiltshire equations And JAX-based simulation It has been used, and has proven to be able to comfortably handle an 81-satellite cluster at an altitude of 650 km.
  3. Radiation Shielding for TPUs
    Because radiation is so intense in space, TPU chips need to be perfectly processed.
    Google's Trillium TPU v6e In a proton beam test, 15 krad(Si) It has been proven to be radiation-resistant because it was able to withstand the impact. 

 

Economic Viability – Decreasing Launch Costs 

Space-based AI was previously not possible and was not cost-effective due to launch costs.
However, Google estimates that it could fall below $200/kg even before the 2030s.
At that point, space-based AI could reach a level where it can be used at the same cost as an Earth data center. 

Future Plans – 2027 Prototype Mission 

Google has partnered with Planet Labs 2027 The Prototype Satellite is planned to be launched twice this year.
In this mission 

  • Testing TPU performance in space 
  • Optical link ML sharing will be tested. 
  • We will study radiation and thermal effects. 

In the future, we aim to build entirely solar-powered “AI constellations.” 

Why Project Suncatcher Matters 

Aspect  Impact 
Sustainable Power  Continuous energy generation with solar power 
Scalable Compute  Unlimited expansion on satellites 
Environmental Benefit  Reducing anomalies on terrestrial data centers 
Engineering Innovation  Creating new optical networking and radiation resilience 
Long-Term Vision  Launching a new space-based AI infrastructure 

 

Summary 

Project Suncatcher is Google's next step back into the "moonshot."
This project will take AI computing to the next level by harnessing the power of the sun and the vastness of space. 

Like quantum computing and self-driving cars — this project could become a prestigious experiment that could redefine the future of technology.
If successful AI systems can be powered by solar power and advanced computing networks outside of Earth. It will become a new way to use it.