By Thor Olavsrud
Aug. 30, 2016
Accurate and hyperlocal forecasts are even more important for drones than they are for manned aircraft, Miller says.
"These drones are very susceptible to weather," he says. "Rain can have an impact; snows, wind is a major impact. Extreme temperatures."
"Most airline operation occurs above 10,000 feet, outside arrival and departure activities," Miller says. "Drones are typically in the lower 500 feet of the atmosphere. The weather and meteorology changes rapidly in that lower boundary layer. You encounter a lot more local weather — very localized weather phenomena that can affect your operation. Those are particularly challenging to forecast."
The Weather Company's forecasting platform already produces precise weather forecasts every 15 minutes for 2.2 billion locations worldwide. It uses machine learning and more than 100 terabytes of third-party data daily, and a network of more than 200,000 personal weather stations that report in minute-by-minute. It's also starting to work on ingesting air pressure data gathered from cell phones and instrumentation data from aircraft, which can help it analyze wind speed and turbulence data, which can then be deployed to other aircraft.
Drones add to big data
Soon, sensors on drones themselves will begin adding to the data trove, along with sensors from smart home technology and connected automobiles.
"The technology that sits behind this forecast engine is notable," Miller says. "We're leveraging the cloud to produce on-demand forecasts with great precision. We're leveraging not only the best science, but also broader sets of data from the Internet of Things."
"What we're fundamentally doing is mapping the atmosphere," he adds. "We're really combining a lot of these technologies and applying them to real-world problems."