Microsoft's AirSim adds self-driving vehicle simulation tools

AirSim launched a car simulation feature to help research and development of self-driving vehicles. Microsoft's open source AI simulator is available for autonomous vehicles by providing a realistic environment, vehicle dynamics and sensing to examine how operate safely in the open world.

Earlier this year, Microsoft's Aerial Informatics and Robotics as a system to test artificial intelligence made additional features for the development of autonomous vehicles. The latest AirSim version is available on GitHub as an open-source cross-platform offering.

Tinuku Microsoft's AirSim adds self-driving vehicle simulation tools

"AirSim is also making additional tools to test air vehicles. We've made it easier for people to simulate unmanned aircraft by adding built-in fly controllers," said Principal Research Software Engineer at Microsoft, Shital Shah.

The new equipment enables quick experiments in control and estimation algorithms without the need for costly debugging and development. New versions include car simulations, new environments, APIs for easy programming and ready-to-use scripts to start research.



AirSim comes with detailed 3D urban environments that cover a wide range of diverse conditions, including traffic lights, parks, lakes and construction sites. Users test the system in several types of environments, including urban, semi-urban, vegetation and industrial environments.

The simulation contains more than 12 km of roads that span over 20 city blocks and has been developed as a plugin for "Unreal Engine" like a popular tool for game development. AirSim also allows researchers and developers to incorporate new sensors, vehicles or even use different physics engines.

Comments