Greetings. The Open Metaverse Interoperability Group is a volunteer-driven community of people interested in what might still more often be called cyberspace if a particular work of science fiction didn’t get so popular.
They are particularly interested in facilitating interoperability via open source means in order to prevent the vision of a shared virtual world system from being captured by any single organization that might wish to use a walled garden to their advantage at the expense of the public.
Recently I decided to check in on how ROS was doing as a project, and considered that given how both organizations take on serious challenges of interoperability in computing, there might be some areas where they can cooperate toward a common goal. I’m not entirely sure on the specifics, but some general areas of overlap that I suspect would be in basically the order I thought of them:
Open-RMF is concerned with managing robotic traffic between vendors, which involves mapping environments, and I guess there might be some niche uses for stuff like using an augmented reality headset to project expected paths of robot traffic onto the environment, or something similar but for overlaying that information onto the camera feed of a remotely operated device. Taking a glance at RMF-Web’s page, it seems to be working great, so I don’t know if there’s a pressing need for an alternate client right now, but the option is there if it ends up being convenient at some point.
Secondly, the fact that the ROS communication graph is based on publish-subscribe messaging leads me to think that there could be some idea on high-level architecture between robots and virtual world agents, given that publish-subscribe messaging it already a widely accepted thing in internet development, so there’s be a bit of a common language for developers. Third, I’m thinking that if OMI’s GLTF extension group’s work with physics and scripting keeps going as well as it is, it might eventually be useful to use Gazebo as a reference for a web-based simulator, so that it might be able to export designs to or import them from the existing Gazebo web application.
As I’m sure my tone already conveyed above, I don’t want to sound like I’m urging people to rush off to implement things specifically on my behalf. So far I’ve mostly just listened in on some of OMI Group’s Discord voice calls and occasionally pointed out where very simple things like RSS feeds might be applicable. OMI Group is in a continual process of defining itself, with a fairly wide scope, so if anything I hope that keeping it in informal contact with related groups could help avoid unknowing duplication of work. I’ll be linking this post in OMI’s Discord, and look forward both to any further discussion that comes from it, as well as hopefully renewing my interest in robotics in general in the near future.