The intersection of software and hardware is crucial in driving advancements in robotics and artificial intelligence (AI). Canonical and NVIDIA are collaborating to drive this synergy. This collaboration promises to enrich users of NVIDIA Jetson Orin hardware with the versatility and stability of Ubuntu.
In this blog, we’ll delve deeper into the specifics of this collaboration for the NVIDIA Jetson Orin family, the recently announced availability for Isaac ROS and the availability of optimised Ubuntu images for NVIDIA Jetson Orin based systems for early-access partners.
Ubuntu and ROS on NVIDIA Jetson Orin
The availability of an optimised Ubuntu kernel and userspace with NVIDIA Jetson Orin will mark a significant milestone in the robotics and AI domain. For about a decade now, Ubuntu has been the operating system of choice to develop solutions using ROS. NVIDIA Jetson, the leading-edge AI and robotics platform, will soon be able to use Ubuntu’s adaptability and development environment. Although NVIDIA JetPack has been based on the Ubuntu packages for a while, it was not possible to get kernel security updates, Ubuntu certification or support for real-world deployments.
Soon, support for the full line up of NVIDIA Jetson Orin modules will come to Ubuntu, providing customers long-term support (LTS) and security patching on Ubuntu 22.04. Canonical is committed to regression testing and validating periodic updates to the kernel, including CVE patching, to ensure robustness and stability. Keeping the kernel and operating system up-to-date will ensure deployments in the field are immune to increasingly sophisticated attacks.
Further, for roboticists – both enthusiasts and enterprises – the integration isn’t merely about having Ubuntu run on NVIDIA platforms. It is also about ensuring that the rich ecosystem of ROS is optimised to tap into the hardware acceleration that NVIDIA offers to the community. By doing so, roboticists can ensure the highest performance and efficiency in real-world scenarios.
Having a validated operating system, with optimisations on ROS leaves more bandwidth for developers to focus on developing real applications.
Isaac ROS reaching a new milestone for the ROS community
The NVIDIA Isaac Robotics platform is designed from the ground up to advance AI in robotics. The latest Isaac ROS updates make NVIDIA AI perception easier to leverage for ROS developers. And the latest release of Isaac Sim is packed with new features to easily develop, test, and train AI robots in the virtual world before deploying them to the real world.
Isaac ROS, NVIDIA’s robotics software framework, has now matured into a production-quality tool with its first General Availability (GA) release, ready to be used with NVIDIA Jetson Orin. One of its standout features is specifically supporting ROS 2 Humble packages. NVIDIA will establish a build farm and compile ROS 2 Humble packages for Jetson, thereby sparing developers the often laborious process of compiling from source and streamlining development.
Furthermore, a key advantage is the optimised bridge that Isaac ROS has to ROS 1. This allows developers to use powerful Isaac ROS features in ROS applications and makes it easier for them to switch between ROS versions. The emphasis on bridging the gap between versions is further highlighted by the integration of the Stereolabs ZED camera.
Lastly, a noteworthy addition to the Isaac ROS toolkit is the introduction of the Centerpose 3D pose estimation DNN. Leveraging NVIDIA Isaac Transport for ROS (NITROS) acceleration, it provides enhanced processing speeds and accuracy in estimating the 3D pose of objects, making it a valuable asset for developers working on robotics applications requiring precise object orientation and positioning.
Get early access to optimised Ubuntu images for NVIDIA Jetson Orin
Although previous Jetson generations, up to and including Xavier, lacked a certified Ubuntu solution, the tide is now changing with the Jetson Orin family. Earlier this year, NVIDIA and Canonical announced our collaboration to enable NVIDIA IGX Orin and DRIVE platforms.
This collaboration has also benefited the broader Jetson Orin family, with Ubuntu images tailored for NVIDIA Jetson AGX Orin, expected to be available later this quarter for early-access partners.
These new Ubuntu images leverage the same NVIDIA-optimised kernel as their higher-end industrial and automotive cousins. Over time, as NVIDIA user space content continues to be integrated into Ubuntu, these images will be the bedrock of future customer solutions. Ubuntu support for the broader Jetson Orin family, including Orin Nano and Orin NX, will also be catered to, providing developers and enterprises with options based on distinct use-cases and needs.
If you are interested in getting access to the early-stage Ubuntu images for the NVIDIA Jetson AGX Orin developer kit from Canonical, please reach out to us.