The new Microsoft to train drones through 3D simulations
Microsoft takes up the training of drones and other autonomous mobility vehicles through the use of simulation under 3D environments with the launch of Project AirSimarriving today Monday in limited preview phase.
The announcement took place at the Farnborough International Air Show, and defines it as “a new platform running on Microsoft Azure to safely build, train, and test autonomous aircraft through high-fidelity simulation.”
Taking inspiration from the original AirSim platform
Perhaps some may remember that it is not the first time that Microsoft has carried out simulation training for drones and other autonomous mobility vehicles, since in 2017 it launched the open source platform AirSim, which ended up shelving years after experimentation.
Although both platforms are based on the use of Artificial Intelligence, one of the main differences is that the original platform required a much higher level of coding and machine learning expertise, while Project AirSim makes it all easier.
For this, not only offer a library of diverse environments of urban and rural landscapes in 3D but the Azure platform will be responsible for generating “massive amounts of data to train AI models on exactly what actions to take in each phase of flight, from takeoff to landing” , and will have sophisticated pre-trained AI models “to help accelerate autonomy in aerial infrastructure inspection, last-mile delivery, and urban air mobility”according to the company.
Project AirSim is oriented to training from domestic drones to passenger transport aircraft, among others, including drones for the transport of goods, for the inspection of objects, and more.
For Gurdeep Pall, corporate vice president of Business Incubations in Technology and Research at Microsoft:
It comes as advances in artificial intelligence, computing, and sensor technology are beginning to transform the way we move people and goods.
For Balinder Malhi, Project AirSim Engineering Lead:
We built Project AirSim with the key capabilities that we believe will help democratize and accelerate aviation autonomy, namely the ability to accurately simulate the real world, capture and process massive amounts of data, and code autonomy without the need for extensive aviation experience. AI
Those interested in taking part, as it is a platform in limited preview at the moment, must fill out this web form to find out about all the news that comes along the way and choose to enter it.
More Info/Image Credit: Microsoft
