My issue with the openmower one is that I don't know how to solder and I'm too afraid I'll break some very expensive components, so it would be nice if I could buy something mostly pre-built.
As for repalmaker, I worry not having GPS won't work since my garden is quite large and a robot bouncing around randomly would probably never finish, I also have a separate front and back (there's a path between them) and I'd like the robot to finish one area then, go to the other area by itself.
Torn. This is clearly a cool idea, and it would save a lot of work (which I would immediately offset with all the work putting it together) But then it will be that much easier to keep a traditional lawn going, which I'd love to get away from over the long term.
That's a great point I didn't even think about. It would be a lot of work up front but it would be much cheaper and less maintenance for me to dig up the grass and plant native plants like clover.
The biggest hurdle I've ran into is that GPS resolution just isn't accurate enough. If you want something that can follow a planned route without having to rely on a bunch of obstacle avoidance (which is difficult enough in itself) then you'll have to set up some sort of local RTK in conjunction with the GPS.
Currently I'm just setting my mower up like a FPV drone that I can drive around remotely from the comfort of air conditioning. Hoping someone comes out with a good automated solution.
Ardumower is the one I've always intended to roll with. Eventually lol. The original nav method is a buried perimeter wire, but they do offer a GPS-RTK kit.
Just because the software is open source doesn't means a product running it is going to be free. Heck, even some FOSS projects have financial contributors who get perks like software support, access to pre-released software, and input into feature development.