Platform to accelerate autonomous flight – Microsoft AirSim

Josh Riedy knew it wasn’t real – that he wasn’t actually hovering near the top of a wind turbine in North Dakota, hundreds of feet off the ground. But it didn’t matter when he looked down. His stomach still dropped as if he were on a rollercoaster.

The CEO of Airtonomy was inside a digital replica of a real wind farm at the time, trailing a simulated drone through virtual reality glasses as it inspected towering turbines. His North Dakota-based company has been using these hyper-realistic simulations to train autonomous aerial vehicles that are now inspecting wind farms, surveying wildlife and detecting leaks in oil tanks across the Midwest.

Every one of those AI-powered flights first happened countless times in simulated 3D worlds. Because if AI is the key to building autonomy in the air, data is the key to building AI – data that is impossible to get in the real world, Riedy said.

“You don’t want to fly drones into wind turbines, powerlines or really anything for that matter,” Riedy said. “Coupled with the fact that winter can literally last 7 months in North Dakota, we realized we needed something other than the physical world to design our solutions for customers.”

The answer was Microsoft’s Project AirSim, announced today at the Farnborough International Airshow. Project AirSim is a new platform running on Microsoft Azure to safely build, train and test autonomous aircraft through high-fidelity simulation.

In these realistic environments, AI models can run through millions of flights in seconds, learning how to react to countless variables much like they would in the physical world: How would the vehicle fly in rain, sleet or snow? How would strong winds or high temperatures affect battery life? Can the drone’s camera see a turbine’s arms on an overcast day just as well as a clear one?

Project AirSim uses the power of Azure to generate massive amounts of data for training AI models on exactly which actions to take at each phase of flight, from takeoff to cruising to landing. It will also offer libraries of simulated 3D environments representing diverse urban and rural landscapes as well as a suite of sophisticated pretrained AI models to help accelerate autonomy in aerial infrastructure inspection, last-mile delivery and urban air mobility.

Project AirSim is available today in limited preview. Interested customers can contact the Project AirSim team to learn more.

Article source: Jake Siegel, Microsoft

Other articles