Robot lawnmowers - boundary wire detection

[larkim]

Recently got a dose of laziness and bought a robot mower. Works really nicely

One thing I haven't got clear in my head is how it knows whether it is inside or outside of the charged boundary wiring.

I can power down the whole system, move the robot to just outside the boundary loop, and when it is all powered back up it instantly detected as being outside the wire.

Anyone know how this works? What goes down the low voltage cable that is the boundary wire to allow the robot unit to detect where it is in relation to being inside or outside?

[Jeff Layman]

Have a look at the article here:

In particular, see the section "Sender/receiver principle using polarity change ("Perimeter v2")" which appears about two thirds of the way down. Quote: "While crossing the perimeter loop, something interesting happens: signal changes polarity, that means negative and positive voltages reverse each other. By using this principle, crossing the perimeter wire will be detected. The robot knows its current perimeter state all the time (inside/outside)."

I'm not sure if crossing the wire is the detection for inside/outside, or simply being outside is different from being inside - it's a bit ambiguous. To my way of reading it, however, it tends towards always knowing where it is, whether it actually crosses the wire or not.

[Martin Brown]

When inside the boundary loop it gets a slightly better signal and there is a null and then a phase reversal as it crosses the boundary wire. It isn't dissimilar technology to a current based hearing loop. Except that the lawnmower can afford to use a fair sized loop antenna.

[NY]

I think the crucial part of the OP's question was what happens if the mower is turned on outside the loop. IN this case, it has no phase-reversal event to detect when it is going outside. If the mower can detect that it is

*already* outside (as opposed to having just *gone* outside) that suggests that it knows what the normal phase of the signal is, to know that it is receiving a 180 degree phase-shifted version of it.

[The Natural Philosopher]

Depends if it also receives something else...a sort of synchronisation clock via e.g. radio.

There are also algorithms which would allow it to determine where it is by shuffling a bit.

Inside the loop field strength would be relatively constant but outside it would be tailing away rapidly..

[Dave W]

I know nothing about this, but I have a theory. If there was DC current in the loop (of course there isn't), inside the loop might look like a N pole and outside would look like a S pole.

If instead a +ve pulse is fed to the loop followed by a -ve pulse, with a gap before the next pair, the robot mower knows what to expect when inside the loop, but would get the opposite effect outside, so knows which way to move.

[Adrian Caspersz]

Somewhere in the world of patents, you'll find the answer ...

Automated lawn mower - US5974347A

... but you may be reading through that for weeks...

[The Natural Philosopher]

THAT is pretty cunning. I think you have nailed it.