[andrewtbham]

That is awesome. I have wanted to do this for an electric mower. And maybe eventually train it to drive itself with ML. You could even start a service to cut yards and use the public records of the lots to set a boundaries for the yard.

[Cthulhu_]

Have you looked at what's available on the market? I don't know if they still need a wire around the lawn to set the boundaries these days though.

That said, these things doing the rounds around a neighbourhood would be cool. But then, so would street sweepers and weeders.

[tstrimple]

Have you seen OpenMower?

https://github.com/ClemensElflein/OpenMower

[toomuchtodo]

https://www.scytherobotics.com/

[samatman]

There's an Ardumower project, I'm intending to work on building one over the winter.

I dream of far more sophisticated home garden robots, but would settle for farming out the basic haircut to a robot.

[bradstewart]

I would definitely buy a Roomba for my lawn (assuming it worked well). Obviously more safety concerns than a vacuum cleaner, with the spinning metal blades and all, but seems doable.

[fennecfoxen]

Existing lawn robots still mostly rely on a buried cable to form a boundary-line. The next generation vision/lidar/gps systems are clearly in development but not commercially available in the US (the Segway NaviMow is available in Europe).

[ftrobro]

Husqvarna EPOS does not need physical boundary wires and seems to be available in the US too:

https://www.husqvarna.com/us/discover/epos/

[samatman]

This year was the tipping point for the US market, they weren't readily available until late summer. I'll have a (ground) wireless lawn robot next year one way or another, it would bring me more satisfaction to build one but that doesn't mean I'll actually do it.

[fennecfoxen]

That's new — but those are still $5000 models "designed for professional fleet use", a little on the steep side. I'd give it a little longer for that to percolate downmarket towards the likes of the Navimow which is around the €1500 level.