News for hackers. Stuff that matters.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by userbinator on Thursday April 16, 2015 @10:56AM (#9388323)
I'm not sure why they're so obstinate about using a very-high-frequency band of RF for this. Is GPS not accurate enough? How about a more mechanical method, or near-field communications?

the only way to get them to work is to dig a trench along the perimeter of a lawn and install a wire that creates the electronic fence needed to ensure the automaton don’t wander beyond the property line.

If they really want to implement that method, shopping cart locking systems have done it before, and they don't need 6GHz+ RF:

http://www.wired.com/2007/07/hack-remote-sho/

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by pjc50 on Thursday April 16, 2015 @11:10AM (#9388453)
GPS accuracy on its own is nowhere near good enough. Differential GPS can give you 1cm accuracy: http://www.blackroc-technology.com/technologies/gnss/dgps/
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by 0xffff2 on Thursday April 16, 2015 @11:14AM (#9388487)
It seems like installing the base station for differential GPS would be easier than installing the electronic fencing descried in the article.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by pjc50 on Thursday April 16, 2015 @11:17AM (#9388518)
Yes, although it may be more expensive (I suspect you have to license a whole pile of patents)
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by oh_sigh on Thursday April 16, 2015 @11:35AM (#9388667)
Easier, maybe? But it would cost tens of thousands of dollars and be subject to even more regulations, so I don't think that is a good solution.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by 0xffff2 on Thursday April 16, 2015 @01:01PM (#9389232)
Why would it be subject to more regulations? You just need two GPS receivers and a way to link them (e.g. WiFi).
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by oh_sigh on Thursday April 16, 2015 @09:57PM (#9391890)
Sorry, but I don't think you know how differential GPS works. It is not just using two GPS devices and averaging their location. Wikipedia is a good place to start: http://en.wikipedia.org/wiki/Differential_GPS
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by mindslight on Thursday April 16, 2015 @11:18AM (#9388531)
I've got to believe there's good reasoning from iRobot, even though none of the articles I've seen have covered it (tsk tsk IEEE Spectrum). Perhaps wanting to bury the transceivers rules out 2.4GHz and 5GHz, while needing tight directionality and small antennas rules out the lower stuff?

Having said that, radio interference is taken pretty seriously (eg http://en.wikipedia.org/wiki/United_States_National_Radio_Qu...). The free-for-all that we're familiar with is really just a few tiny consumer bands.

DGPS seems like a better idea (even easier to install a virtual border by manually driving the robot around the edge), but would probably require some above-surface beacons. I'd imagine iRobot is trying to cater to the type of people who think grass needs to be mowed every week and wouldn't want even a few beacons sticking up.

Still I'm sure there's another way. And if iRobot wants to win in even just the court of public opinion, they have a steep argument to make.

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by ryandrake on Thursday April 16, 2015 @11:21AM (#9388552)
GPS isn't enough. A lot depends on the receiver, antenna, how many satellites are in view, and what's going on with the atmosphere, but unaugmented GPS is generally considered accurate to about 15 meters. Not good enough for a lawn unless you're in a rural area or your neighbors don't care if you cross the property line.

GPS in your car only appears to be more accurate than that because they're typically "snapping" the GPS signal to an underlying road map.

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by rwmj on Thursday April 16, 2015 @11:06AM (#9388418)
Tomorrow's World, 1979: https://www.youtube.com/watch?v=Ade1JYpscl0

I wonder if the real reason is that they can't get a patent for simpler methods because it's all been done decades ago?

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by TeMPOraL on Thursday April 16, 2015 @12:16PM (#9388957)
I so wouldn't be surprised if that would be the actual reason.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by rbarris on Thursday April 16, 2015 @05:36PM (#9390883)
I would wager they are using UWB for position tracking, as seen in the DecaWave devices, which can operate in the 6GHz spectrum. No GPS required, but you do need multiple nodes to range against.

http://www.decawave.com

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by cwyers on Thursday April 16, 2015 @11:06AM (#9388421)
> Is GPS not accurate enough?

It really isn't.

Re:The Roomba for Lawns Is Really Pissing Off Astronomers by teamonkey on Thursday April 16, 2015 @11:10AM (#9388449)
Or optical, even? Once you've accepted that there will need to be a stake in the ground, you'd already eliminate a lot of the line-of-sight problems.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by pmorici on Thursday April 16, 2015 @11:03AM (#9388389)
You mis read it, they want to _avoid_ having a buried wire system. They are instead proposing a stake system which is what they are requesting a waiver for.
Re:The Roomba for Lawns Is Really Pissing Off Astronomers by hackmiester on Thursday April 16, 2015 @11:11AM (#9388462)
Your quote is of exactly the method they do not want to use.