Open Robotics Technology Strategy for 2026
Context and organisation mission
The OSRF’s mission is to support the development, distribution, and adoption of open source software for use in robotics research, education, and product development. Within this mission is the need to continuously develop and maintain software and services that meet the needs of our users, as those needs change.
Aspirational goals
The OSRF believes that the following goals are our guiding navigation markers. They capture what we believe our users most need from the software and services we provide.
Provide trustworthy solutions for robotics
Using Open Robotics software is worry-free. When you build a product using an Open Robotics project’s software, you know what is in that software, that the software will do what it claims, and that the software does what you need. You will not be surprised by how the software behaves and it will be able to meet your performance requirements. You will not be surprised by a sudden licensing dispute, or the loss of an important supplier. You know that you will be able to respond to future changes because the software gives you the flexibility you need without incurring an outsized cost.
Enable agility in robotics development
Open Robotics software enables users to build robots and robot applications quickly. It gives development teams the flexibility and capabilities they need to iterate rapidly, while supporting the growth of the team and the systems they are building. It gives individuals the ability to leap quickly towards building the application they are interested in, without needing to worry about the foundational functionality they want to build their application on. As new technologies become available, Open Robotics software makes it possible for a robot application to be adapted to take advantage of that new technology. As customer needs change, Open Robotics software allows the provider of a robotics solution to pivot to meet the new requirements, and know that they still will be able to provide a robust and trustworthy solution.
“We build our product on top of Open Robotics software because we know that it will meet our needs throughout the product lifecycle, from exploration in the lab, to first prototypes in the field, to scaled production deployment.”
“Technology in robotics is moving fast, and we need to be able to keep up. Using Open Robotics software allows us to explore solution options quickly, to build reliable products and services rapidly, and to deploy and maintain with speed, without wasting resources.”
Strategic priorities
The strategic priorities of the OSRF give guidance to our projects and contributors. They describe what we consider to be the priorities for development through one calendar year. The OSRF strategic priorities for 2026, covered in detail below, are:
Enhance support for Physical AI,
Expand accessibility and ease of use, and
Adapt modern tools for production systems
Projects use these priorities to guide their roadmaps. Contributors, including new contributors looking for somewhere to make an impact, can use the priorities to find something to contribute.
Enhance support for Physical AI
The new capabilities offered by recent advances in neural networks hold great promise for robotics. The drive to produce and use new Physical AI systems is strong. The OSRF will provide the premier open-source platform for developing AI-driven robotic systems.
Runtime efficiency of data transfer is commonly seen as a key measure of how suitable a platform is for use in Physical AI. We argue that even more important is developer efficiency - and as it has always been, this is a driver of our design decisions. ROS has historically been amongst the best platforms for onboarding new people into robotics and into your specific application. It is not yet at this point for Physical AI and machine learning experts. Our projects need to support how experts in Physical AI want to work, the ways in which they want to manipulate data, and avoid introducing slow-downs into their workflows. Where slow-downs in data transfer exist, they will be measurable and quantifiable, so engineers can make informed decisions about when throughput and latency is a problem.
The Open Robotics Suite is also in a unique position to bring Physical AI to the edge. Our projects will support the needs of Physical AI when it comes to interacting with hardware by providing appropriate deployment mechanisms that include low-level embedded systems as well as high-powered computers.
Expand accessibility and ease of use
To support agility in robotics development, the Open Robotics Suite must support rapid installation, rapid configuration, and ease of understanding the platform. Only with these can a robotics engineer quickly test potential causes of unwelcome behaviour, try new ideas, comparatively evaluate approaches to solving a problem, and test new technologies as they arise. In the same way that a DevOps engineer will spin up a cluster to test an idea for the afternoon, the Open Robotics Suite must enable a robotics engineer to quickly put together a robot system or even a fleet of robots, whether in simulation or on ready-to-go hardware, for a quick test of anything that needs evaluation. This involves all aspects of our platform, from installation through to the APIs and tools used.
In addition to our existing installation methods, with their focus on specific known platforms, our projects will support installation via package managers and installation mechanisms — such as modern CMake, recent Python installation tools, and Pixi — that are specifically designed to enable “simple” installation or that are flexible in platform support.
We are undertaking a significant new effort to improve our documentation, and plan to continue this work throughout 2026. New documentation structures and additional documentation, with a renewed focus on those who need documentation the most, will improve the accessibility of the Open Robotics Suite throughout the user community.
Adopt modern tools for production systems
The Open Robotics Suite has a long history. Over that time, it has developed its own ways of doing things. However, in parallel other groups have also developed new tools and approaches. Sometimes these are better than what the Open Robotics Suite uses or provides for some of our users. Allowing use of new programming languages such as Rust, supporting new infrastructure tools such as Bazel, and providing new assets such as SBoMs promise to enable new levels of reliability and trustworthiness in the Open Robotics Suite - and help us meet contemporary expectations from professional users and from new legal regulations. Through this, we will enable our users to use modern practices, such as producing and taking advantage of hermetically reproducible builds and using modern software sources, while maintaining the traditional federated approach of ROS.
Through our implementation of these three priorities across the whole of the Open Robotics Suite of projects, we will continue to move the platform forward so that it continues to meet the needs of robotics, even as this field goes through significant technological changes.