Microsoft’s AirSim project to test safety of self-driving cars

Realistic virtual environment for assessing AI systems to be provided open source

AirSim is an open source research project from Microsoft that provides a realistic simulated environment for testing the safety of artificial intelligence systems. Originally designed to run simulations for quadcopters and other drones, it's now being extended to test self-driving cars and other autonomous vehicles.

The virtual environment created by AirSim is both visually and physically realistic, providing urban landscapes complete with traffic lights, parks, office blocks, and city-centre areas; it has over 12km (7½ miles) of drivable roads spanning more than 20 city blocks.

Microsoft is providing the technology open source to encourage more researchers to enter this space.

"It requires infrastructure to build expensive hardware platforms, large amounts data and the ability to quickly test and benchmark results. We aim to make these various aspects of developing self-driving cars available to a broader group of researchers by providing an open, community-driven platform for testing those algorithms," Shital Shah, principal research software engineer and Ashish Kapoor, principal researcher, explained on the Microsoft Research Blog.