Auterion driving ROS 2 adoption for flying robots

December 9, 2020

Auterion announces a comprehensive Skynode integration with open source robotic libraries and middleware at ROS World 2020

The Robotic Operating System (ROS) provides libraries and tools to help software developers create robot applications. ROS features include hardware abstraction, device drivers, libraries, visualizers, message-passing, and package management capabilities.

Since ROS began in 2007, there have been significant changes in the robotics and within the ROS community. We now have a massive number of developers driving an ecosystem of thousands of different packages, from simple camera drivers, to complex SLAM and NN-based control and navigation algorithms – all open for use and exploration by anyone in our industry.

Requirements for a robust and flexible middleware architecture, enhanced security and interfacing have led to several recent updates on the core implementation of ROS. Today, the goal of the ROS 2 project is to adapt these changes with a revamped architecture built on top of the Object Management Group’s (OMG) Data Distribution Service (DDS).

To avoid deprecation of the current applications developed with ROS, ROS 2 is not being developed as a version update. Rather, the new software runs alongside the existing version and does not replace it. The latest LTS release, Foxy Fitzroy, has reached a state of maturity that will provide a faster and more consistent adoption by our industry.

Auterion expands its support for ROS 2

The Auterion software platform is a combination of a flight controller, an operating system running on the mission computer that lives on the drone, a ground control station, and the Auterion Suite running in the cloud.

The innovative engineers at Auterion are proud to lead the adoption of ROS 2 in aerial robotics platforms by binding the powerful capabilities the Auterion Enterprise PX4 firmware internals with the cutting edge robotics libraries and tools provided by ROS 2. This tight integration with the PX4 internals is made available by the micro RTPS bridge. Developed in 2017 to bridge between PX4 and the DDS world, it leverages eProsima’s serialization libraries, as well as Fast DDS (formerly Fast RTPS), eProsima’s C++ implementation of DDS. Since then, and with several improvements provided by Auterion over these years, it has allowed for a more flexible interface with the control, estimation and navigation modules of PX4 – without the need of a secondary protocol, like MAVLink, and the burden of translating the required microservices.

Last month at ROS World 2020, the premier annual developers conference, Auterion participated in several live technical talks and panels during the virtual show live streamed across the world. We also announced our product roadmap to bring ROS 2 and DDS to our integrated drone avionics Skynode platform.

Robotics is the intersection between hardware and software. And until now, there hasn’t been a good platform that you can do both at the same time. Skynode provides this by offering standard hardware and standard software that you can use to solve your robotics problems. This powerful solution allows you to simultaneously run complex low-latency robotics control algorithms as well as high bandwidth sensor data processing on a Linux-based operating system.

Soon, customers wanting to add cutting-edge robotics features to their drones, such as path planning, non-linear decision making, visual odometry, and obstacle avoidance, will be able to start developing robotics applications using Skynode. This is accomplished through a powerful combination of our enterprise SDKs and ROS 2 APIs and tooling.

Auterion is the industry’s dominant force that supports the most powerful robotics software and infrastructure on drones and in flying robots that are based on open standards and community-driven development. We are excited to announce we will be integrating our Skynode enterprise product with the most used open-source robotics libraries and middleware provided by ROS 2, along with the most used open-source autopilot ecosystem, PX4.