How can you ensure that autonomous drone flights are safe in the future without first having to make mistakes in real life to learn from them? By simulating such drone flights. That is the premise of Project AirSim, a simulator for autonomous drone flights. The simulator is an initiative of Microsoft.

Train (automatic) pilot with Project AirSim

Soon, Artificial Intelligence (AI) is expected to play an important role in autopiloting drones. But how do you test the behavior of an autopilot in extreme conditions such as rain, wind, or snow? How do you determine the exact influence of wind on flight time? Can a drone make good recordings for inspections on a foggy day?

Project AirSim should provide the answers to these and other questions. The idea is that computer models of drones can perform the same flight millions of times in a short time, but each time under slightly different conditions. This way, the AI ​​autopilot can be trained to act in deviant situations….