FlytOS: First stable release is out!


After months of extensive testing, we are happy to announce the first stable release of FlytOS, the framework for advanced drone applications.

As some members of the community would recollect, FlytOS is built on ROS and Linux, to help developers quickly integrate high-level applications, such as, stream live-video, computer vision, machine learning, mobile-app based drone control, cloud connectivity for drones, etc. FlytOS is now compatible with a wide range of companion computers, such as, Odroid XU4, Raspberry Pi, Nvidia TX1, Intel Edison, Intel Aero and Qualcomm Snapdragon Flight; so, whatever apps you build, they are automatically compatible with all of them. 

FlytOS is preconfigured with a generic development environment, that includes your favourite libraries, such as, ROS, OpenCV, GStreamer. Focused on developers, FlytOS provides extensive APIs in ROS, Python, CPP, RESTful, websocket to interact with drones. Further, apps built for FlytOS can be tested using FlytSIM, a 3D simulator, based on ROS/Gazebo. FlytOS, of course, comes loaded with a number of sample codes to help you get started quickly.

Here is a sample application, where we use FlytOS to implement machine learning algorithms on TX1 to automatically detect and classify objects (everything is processed in real-time on the drone): 

Next application demonstrates visual servoing of a gimbal mounted camera, using FlytOS:

And one more, for visual object detection, tracking and following:

To get started, just flash a memory card with the FlytOS image and plug it into your companion computer.

Get FlytOS here:

Detailed documentation is available here:

Checkout the Github repo for sample apps:

For any technical issues, please use the forums:

We hope you will have as much fun using FlytOS for your drone applications, as much we are having building it! Do continue to share your comments and feedback to help us make it better.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • Great work Nitin. Looking forward to trying this out with my RP3 and PXH Mini.

This reply was deleted.