Automotive Grade Linux All Member Meeting Europe 2026

This presentation will provide a status update on the work to create a Xen-based SoDev demonstration in the upstream AGL Unified Code Base (UCB) for the Vibrant Vimba release. The Yocto Project (YP) BitBake multiconfig scheme used to build and assemble the domain images will be outlined, with examples on how it can be extended for more elaborate downstream demonstrations, and potential follow on work will be discussed.

Projects like ELISA and Zephyr have helped move functional safety to the forefront of the open source ecosystem. As a mature project with more than two decades of history, the Xen Project faces a different challenge: how do you evolve an existing codebase, development process, and contributor community to support safety-critical use cases?

This session shares lessons from the Xen Project’s ongoing safety journey, from improving development processes and documentation to producing reusable safety artifacts and collaborating across organizations. Along the way, we’ll discuss the practical challenges, engineering tradeoffs, and community decisions that come with evolving an established open source project.

Whether your project is just beginning its safety journey or is already supporting safety-critical systems, you’ll leave with practical ideas that can help any mature open source community prepare for what’s next.

Dense point-cloud maps are heavy to store, share, and maintain. My research localizes a vehicle against compact 3D building models instead, using panoramic images and semantic cues to match what the camera sees to a lightweight city model.

In this session I explain the approach without heavy math. I cover how I build the lightweight 3D model, how semantic segmentation links image to map, and how the vehicle tracks its pose against simple building geometry. I show where this beats dense maps on size and upkeep, and where it still struggles.

You leave understanding a lighter path to localization, and the trade-offs against point-cloud methods you may already use. I presented the underlying research as a poster at ICRA 2026 in Vienna, titled Semantic Equirectangular Visual Tracking in Lightweight 3D Building Reconstructions, and this talk adapts it for an automotive open-source audience. All results are my own.