Tomorrow.io's radar satellites use machine learning to punch well above their weight

Tomorrow.io, formerly known as ClimaCell, has released the results from its first two radar satellites, Tomorrow R1 and R2, which use machine learning to provide weather forecasting and analysis. The satellites, which are smaller and lighter than traditional weather satellites, have proven to be competitive with larger, more traditional forecasting technology. The company's high-tech approach has shown that their satellites can produce results on par with NASA’s Global Precipitation Measurement satellite and ground-based systems, even with only one radar band and a fraction of the mass.

The company's goal is not to replace existing infrastructure, but to supplement it with a network of satellites that can provide detailed prediction and analysis globally. Their future satellites will be larger and more capable, and they aim to provide real-time precipitation data anywhere in the world. They also plan to use the data collected to train their own and other systems, with more satellite launches planned for this year.

Key takeaways:

Tomorrow.io has released the results from its first two radar satellites, which use machine learning to provide competitive weather forecasting and analysis.
The company's satellites are smaller and lighter than traditional weather infrastructure, such as NASA's Global Precipitation Measurement satellite, and use only the Ka-band radar.
The satellites, Tomorrow R1 and R2, have been able to make accurate and precise predictions and observations, on par with NASA's GPM and ground-based systems.
Tomorrow.io plans to create a global network of satellites for detailed weather prediction and analysis, and is also working on providing real-time precipitation data anywhere in the world.

Tomorrow.io's radar satellites use machine learning to punch well above their weight | TechCrunch

Key takeaways:

Comments (0)

Newsletter