Search Unity

We’ve partnered with Microsoft to bring the autonomous vehicle simulator AirSim to Unity. This collaboration helps democratize the development capabilities for autonomous vehicles and drones by taking advantage of Unity’s performant engine, easy to use C# development environment, and quality content from the Unity Asset Store.

AirSim on Unity

Created by the team at Microsoft AI & Research, AirSim is an open-source simulator for autonomous systems. It’s a platform comprised of realistic environments and vehicle dynamics that allow for experimentation with AI, deep learning, reinforcement learning, and computer vision. With AirSim on Unity, you have the opportunity to create and innovate on an entirely new ecosystem and platform.

“Our goal with AirSim on Unity is to help manufacturers and researchers advance autonomous vehicle AI and deep learning. Unity gives its OEM clients the ability to develop realistic virtual environments in a cost-efficient manner and new ways to experiment in the world of autonomous and deep learning.”

– Ashish Kapoor, Principal Researcher at Microsoft Research & AI

Powerful and performant

Thanks to the relentless focus on performance at Unity, AirSim on Unity offers smooth overall performance running at high frames per second. Paired with Visual Studio, you get the ultimate development environment that’s truly integrated and easy to use.

Cross-platform support

We embrace seamless cross-platform support as a core principle, which is why Linux support will soon follow today’s release of AirSim for Windows. With the upcoming Linux release, you will be able to run highly performant simulations no matter which platform you choose.

Experiment with ML-Agents

Unity’s own machine learning initiative ML-Agents can be integrated into AirSim’s capabilities, allowing for even more experimentation. The open source ML-Agents are available through GitHub and have been positively received with well over 4,000 stars. With the release of AirSim on Unity, the two communities now have a common ground to experiment, develop, and evolve together.

“Using the new AirSim tools, we have trained and evaluated our ML agents for unmanned aerial vehicles inside Unity in mere hours vs. training them in the real world over several days and weeks.”

– Anurag Rana, CEO at Threye

Build environments fast

The Unity Asset Store provides an expansive library of high-quality content that you can use to quickly and easily build complex virtual environments for your simulation.

To get you started, we offer the Windridge City environment as a free download from the Unity Asset Store. This beautiful environment supports both automotive and drone experimentation across urban, suburban, and rural locations. Windridge City is open source just like AirSim, so modify and use it freely.

A big thank you goes out to NatureManufacture and Indago for their efforts in bringing Windridge City to life. They used their own resources and leveraged tools from the Unity Asset Store including “EasyRoads3D” by Unity Terrain Tools and “Gaia” by Procedural Worlds. Another thank you goes out to Rythmos who helped create the AirSim wrapper code. These partners share our passion for advancing autonomous simulation research and embody the quality and diversity of companies in our ecosystem.

Get started with AirSim on Unity

By keeping the AirSim API unchanged, providing sample demo projects, and offering Windridge City as a free download, you can quickly get started with AirSim. Here is a list of helpful resources:

The GitHub repository contains a new Unity folder with the AirSim wrapper code, car and drone demo projects, and documentation. The car and drone projects work with existing sample scripts available in the GitHub repository, including HelloCar.py and HelloDrone.py.

The Windridge City environment is available from the Unity Asset Store and requires Post Processing Stack v2 which you can install via the Unity Package Manager using these instructions.

Note that this initial Windows release is labeled as “beta” with Linux support coming soon. While AirSim supports the core APIs, we are excited to have the Unity community experiment and help us bring out its full potential.

Feel free to share feedback directly in the GitHub repository or comment here on this blog. Happy experimenting!

Deixar uma resposta

Você poderá usar estes atributos e marcas HTML: <a href=""> <b> <code> <pre>

  1. Hello Everyone,
    For my project we need a simulation software, basically the task is a simple pick-and-place (with a robot manipulator), but in outdoor environments.
    The sensors we are gonna use are: Camera, IMU, Torque, Pressure, Force and one different kind of orientation (1D) sensor.

    I tried VREP, it is easy to use and also it supports all the sensors, but the environment rendering is really bad (even the resolution is not good, when streaming from the visual sensor).

    I checked some tutorials of AirSIM and the environment modelling is very realistic with shadows and weather effetcs (which suits perfectly for the project) , however as far I know AirSIM only supports Camera, GPS, IMU and Barometer, also it has no Robot manipulator models (I do know that it’s main focus is on UAV’s).

    So I just wanted to know whether it is at all possible to import urdf model from ROS (as AirSIM have ROS support), also can one model and incorporate these other sensors.

    Any suggestions and advice is much appreciated.

    Thank You.

  2. As soon as I hit play in Unity, the app crashes immediately, with no log that i can find.

    1. Hi Jason,
      Sorry for the inconvenience. This feels like a settings.json issue.
      Probably best if you could post this question to AirSim’s Forum, here: https://github.com/Microsoft/AirSim/issues
      Thanks!

    2. Yeah, you just need to add this line : « “SimMode”: “Multirotor”, » to settings.json
      In Windows this file will be created for you under “My Documents\AirSim\” just after you run one of the demos binaries found over here https://github.com/Microsoft/AirSim/releases

  3. Hello, Do you have a scene for your windridge city that i can use in unity? I’ve gone through all folders and can’t see a scene file. How do i replicate your picture? Thanks

    1. Hi kingz,
      The Windridge City package can be freely downloaded directly from Unity’s Asset Store. The blog had a hyperlink for it on the package name, but it gets a little obscured in a large text like that.
      Here’s the direct link: https://assetstore.unity.com/packages/3d/environments/roadways/windridge-city-132222
      I hope you enjoy it!
      Thanks!

      1. Hi Jose, Thanks for your response. I may not have been clear in my question. I have already downloaded the windridge unity asset. I was asking how to I put the scene you have in my scene. Thanks

        1. Nevermind, i think the asset i downloaded are missing some files. I will re-download it again. Thanks

      2. really Good assets nice work but men i don’t have latest version of unity i have unity 2018 1.1f1 but please don,t delete this assets soon i will download latest version of unity and also i gave you five stars soon by the way your are such a good designer

  4. Oh, one more question: Why not controller support instead of just keyboard?

    1. Hi Jason
      Excellent questions!
      We are trying to keep the more technical questions like yours centralized in AirSim’s discussion forum, here: https://github.com/Microsoft/AirSim/issues
      Would you mind re-posting those there?
      Thanks a bunch!

      1. Yes sir. Will look at that. This is great! Thanks!

  5. Hi, This is great!. A few questions:
    1. Will this run in Unity with the same performance as it runs in Unreal? Or does does the wrapper add some overhead that will slow things down?
    2. Will this support multi-agent simulations for reinforcement training of several autonomous drones at once? For example running batches of 10 or 20 at a time.
    3. I was testing this out and the colliders don’t always prevent the drone from going through the ground. Is this a bug or is there something I’m not setting up right with colliers and collision detection. How do collisions register to Airsim lib? Do you have any optimization tips?
    4. Will this support SITL? If so, can you explain how to set that up?
    5. What is the best place for researchers to post questions like this? (Things related specifically to AirSim on Unity.)

  6. Unity’s going mad…

    1. Hello All,

      As a part of my little project, Im trying to build this simulator on Unity to make it work with MiddleVR. I was having issues earlier since I was dealing with Unreal Engine, but since thats out of the way, I am facing some issues in running the ‘ Cardemo’ .
      To Start with, its mentioned in the GitHub to get Unity 2018.2.15. I was able to build,cmd but when I ran the game it said the game was built on 2018.2.14. So I got the .14 version and ran the car demo but I didnt see any sort of map to play in, just an instance of a floor with the camera settings for the prefabs. Also, when I pressed play, Unity Immediately crashed(Tried it a few times). Anyone facing the same Issue?? I was also thinking of importing traffic simualted data that I created on SUMO Simulator using the TraCi protocol. Is it pssible on Airsim ?
      Regardless, v4z

  7. LOL I recall a forum conversation where I proposed using Unity, Navmesh and Google Maps to get a car to drive you where you wanted!

  8. Why does this talk about ECS and the Job system when neither are production ready or compatible with any of the assets you talk about? The screenshot you include is also clearly not using ECS. Have you been told to throw those terms in to every blog post you do?

    1. Hi Ed,
      Thanks a lot for raising this! My original intent was to list those as examples of Unity’s commitment to performance, but I obviously didn’t phrase it that way.
      That said, whatever you build for your simulation scenarios on this virtual world – traffic, pedestrians, etc – can and should benefit from increased performance coming from leveraging ECS and the Job System. But, as it was stated, and as you correctly caught, AirSim is not taking advantage of that on this initial release.
      Super thanks for the feedback!

      1. Thanks Jose – appologies if I came across a bit harshly!

        1. No worries! You pointed out something that didn’t look right and, with that, helped the blog to convey the right info. It’s a great outcome in my POV and I super thank you!