NASA We chose the weather intelligence company Planette to develop Qubitcast, a quantum-inspired artificial intelligence system designed to predict extreme weather events six months ago.
The purpose of this project is to address current limitations of forecasting technology. This struggles to provide accurate and timely predictions of extreme weather events such as heavy rain, heat waves, and thunderstorms.
Unlike traditional weather modeling systems, which are limited to 10-day forecasting, Planette’s approach combines physics-based models with AI to extend predictions into the future up to one year. The company’s technology uses traditional computers to apply quantum physics principles to identify hidden patterns of Earth’s climate data.
“Planette is one of the first companies to apply these methods beyond theory to weather and climate,” says Kalai Ramea, co-founder and chief technology officer of Planette.
“You can think of it as if you were to read the entire history of the Earth’s systems at once.”
Current long-term weather forecasts rely on simulation models that require a huge amount of computational resources. Qubitcast aims to provide more accurate predictions using significantly less computing power and energy than traditional systems.
“As reliable long-distance forecasting was not available, too many important decisions are made in the dark,” said Hansi Singh, co-founder and CEO of Planette.
“The equation changes the equation by making seasonal predictions more accurate from the subshe equation, but also making it practical to deploy on a large scale.”
The project is funded through NASA’s Small Business Innovation Research Program. This follows Planette’s previous Phase I SBIR grant from the National Science Foundation, which will develop NIVA, the fundamental AI model for Earth System Analysis.
The technology is expected to benefit multiple sectors, including agriculture, emergency management and energy, by providing extended lead times for preparation for severe weather events.
