Michel Hidalgo

@brandonSc @dchw last ping.

> But the protocol itself is nowhere to be found. It can only be reversed engineered. Aha! This may not be 100% true. We shall take a look at some...

Let me try to answer that on behalf of @LucasJSch, a co-worker of mine. > A change like this would keep the graph solver / loop closure logic but be...

Sorry for the delay to reply. Busy week. > This is all pretty tied into Karto and its pretty easy to screw things up in Karto if you're not careful,...

> Maybe its worth us chatting about localization. I hate AMCL and have a roadmap for getting rid of it for something better and could be extended to support 2D...

> This is frankly so much code, I cannot reasonably review it. I'm not really sure what I can promise to be done here. Assuming that LOC count is the...

@SteveMacenski I've reached out to @Shokman to help review this contribution. He's a former co-worker and a friend with expertise in the area and no ties to Ekumen Inc. as...

An efficient implementation of this wonderful piece of research may be an interesting addition and a solution to past issue reports (such as https://github.com/SteveMacenski/slam_toolbox/issues/309). Would you be willing to accept...

> There are sometimes in SLAM sessions "key" nodes that regardless of the information it supplies, is the "backbone" of a loop closure or important tie-in that removing will substantially...

I don't dislike the idea, but it sounds like a lot of scaffolding for a simple tutorial. If we are going in that direction, I'd probably revamp [`beluga-demos`](https://github.com/Ekumen-OS/beluga-demos) for that...