Microsoft Unveils Simulator to Train Autonomous Aircraft Systems
The simulation platform creates drone flight paths and helps businesses prepare for potential challenges
Microsoft has launched its latest addition to the autonomous aircraft industry; a high-fidelity simulation platform called Project AirSim. The platform acts as a drone flight simulator, designed to train and develop the software behind these units and create a library of flight paths. The new simulation platform runs on Microsoft Azure and allows AI models to run through millions of possible flight paths in seconds to identify potential problems – such as dangerous weather conditions or routes along power lines – and how best to respond to those situations. Project AirSim will also store data to create a protocol of simulated 3D environments for future use by autonomous aerial vehicles. Using information from map providers including Bing, the platform can also offer customers 3D views of specific locations. “Autonomous systems will revolutionize many industries and enable many aerial scenarios, from last-mile delivery of goods in congested cities to inspecting downed power lines from more than 1,000 miles away,” said Gurdeep Pall, Microsoft’s Corporate Vice President of Business Incubation in technology. and research. “But first we need to safely train these systems in a realistic, virtualized world.” The platform is based on Microsoft’s AirSim, launched in 2017 – which captures the high-fidelity aspect of simulation while making it more accessible. “We’ve built Project AirSim with important capabilities that we believe will help democratize and facilitate air autonomy,” said Balinder Malhi, Project AirSim Lead Engineer. “This means the ability to accurately simulate the real world, extract and process large amounts of data, and code autonomously without the need for deep AI skills.” Microsoft is also working with industry partners to improve the accuracy of its weather simulation, as well as the sensors the autonomous machine uses to “see” the world.