Building a sensor to locate a burried wire fence

I am doing the old 'Build a robot to cut the grass' bit and while moving the platform abound it not a problem (no cutting motor yet) i have been perplexed about all the different suggestions of how to keep it on the grass. i have read about colour recognition (grass = green), GPS, Laser fencing etc. but i seem to have settled on the idea of a burried wire fence. having done a bit of research i have found someone who has done a similar thing but have no joy in contacting them. Below is a link to their website where they have circuit diagrams of the i2c receiver and a transmitter which sends a FR signal round a perimiter cable.

Receiver <A href='http://www.betuwe.net/~mowbot/UDO5recvRcv.PDF' Target=_Blank>External Web Link</a>
Transmitter <A href='http://www.betuwe.net/~mowbot/UDO5transSch.PDF' Target=_Blank>External Web Link</a>

I like the idea of this unit being i2c as it makes the interface a lot easier but my concern is if i get PCBs made for these using the designs i have been quoted GBP 50 per design (for 4 &amp; 5) little boards respectivly. This seems a lot to lay out on an experiment.

My other problem is that they use a tunned coil to pick up the signal, can anyone advise how i can make a coil which will pick up the correct signal.

thanks

Interesting ! .....and very dangerous if you're not careful !!!

Self powered mower platform?
Irregular shaped lawn?
Flat or inclined?
Is there any margin for error when reaching the boundary?

How about a central post, a self-tightened reel of string connected to the top of the mower platform, and a compass module. Can then work out polar coordinate of platform with respect to centre post. Have internal (memory) map of lawn to be mowed.

Once you've solved this one, you can tackle the vacuuming problem ;-)

Edited by - Jeremy Leach on 11/07/2006 16:24:53

No matter which design you choose, before you commit anything to a &quot;real&quot; professional PCB, build up your circuit and test it as a &quot;breadboarding&quot; setup. This makes for a much more flexible and easy-to-change route (and it's much cheaper, too, and you can re-use the breadboard later for other circuits). Of course this only works well for through-hole components and DIP packaged chips. Surface mount component can often be used in a &quot;dead-bug&quot; configuration - turning the part upside down and soldering little wires on each pin that then connect to the breadboard.

There's virtually no chance your first attempt will be perfect (let alone even work so it makes little sense to spend $$$ on doing a PCB for that. This should be the LAST step when all the issues have been worked out.

Just my 2 cents

Wolfgang

I agree. Breadboarding is ALWAYS the first step. Then your home made pcb. Unless you're rolling in the wonga.

You could dig out your old Physics books to calculate coil diams / no. of winding etc.
And while you have your book open you can get the equation (which we all know boys and girls) for resonant frequencies. And remember the orientation of your coils c/w the buried wire.

Then bench test it.
Got an osillyscope? No? Get one.

I reckon GPS would be a real laugh for this.

To get this good (and safe) is quite a project. Good luck and go for it. But keep your pets indoors. Warn the neighbours.

You'd better do it quickly, otherwise you'll need a PICAXE controlled Combined Harvester.

Edited by - Dippy on 11/07/2006 16:16:53

And most important of all, really consider what can go wrong and how you can both avoid things going wrong and bring everything to a halt if it does.

You are on the verge of this being a safety critical system which would mean that a PICAXE couldn't be used in the project, but without blades it's just another robot, albeit a potentially dangerous one as mentioned.

A mobile unit which could cause injury or death if it got out of control will need multiple failsafes included. A perimeter wire is one thing, but what if the transmitter fails, what if the receiver fails, what if the electrical noise causes the PICAXE to lock-up and not care about whetever it's told or something interferes with readings coming back from sensors ? What about passers by using mobile radio devices which swamp the area with high powered RF ? Will it work perfectly 100% of the time or just 99.9% ? This is a project which you ought to document well so someone independant can really take responsibility for verifying your code and design does what it's meant to.

You will probably require bump switches as a bare minimum which can kill power entirely and are run through an entirely separate system which can still work even if the PICAXE and all else has failed. And it's probably best to make none of the cut-outs auto-reseting without some very overt intervention, preferably one which requires disconnecting the power before it can be reset.

Put the words, &quot;rotating blades&quot;, &quot;children&quot;, &quot;pets&quot;, &quot;blood bath&quot; and &quot;lawyers&quot; in various arrangements and think of the very worse case scenarios you can. I know that sounds like scare-mongering, but risk assessment is a major part of any such project; if it can go wrong, it invariable will, unless of course you've designed for that, when something else unexpected will happen instead

In a rigorously controlled environment it's one thing, but before bolting blades on it and 'setting it free', you really need to do a lot of analysis, experimenting and some soul searching.

Might have to do some toe searching as well!

Here's a nce site that has dealt with most of the mower issues, including the burried wire fence detector.

http://www.robotshop.se/micro/wwwrc_us/indext.htm

They used to have the complete schematics listed, though a quick search didn't produce them. (must have taken the schematyics down when they decided to sell boards).

I have them on one of my computers, but it may be a couple of days before I can find them.

The concept is pretty simple, though. The wire loop transmitter is simply a 555 astable circuit tuned to 38kHz.

The The reciever in the mower is an RC circuit tuned also to 38kHz, then amplified and measured with the COUNT command.

SAFETY--
Several commercial mowers use tilt sensors to disable the cutter if the mower was lifted or jostled. You might consider one of your own

Hi Larry,


&gt; http://www.robotshop.se/micro/wwwrc_us/indext.htm

&gt; (must have taken the schematics
&gt; down when they decided to sell boards).


There is more info, including schematics and coil calculations, in the diary link.

try a trimmer line version instead of blades. it woul be easy to control not to dangerous and an electric trimmer from ryobi is cheap and easy to hack.

This project looks more difficult and expensive than I thought!

Good link Wilf.

I'm so pleased to see that terrible spelling isn't limited to PICAXE forum.

&quot;...not to sell this kit on the US-market since there is always a risk that someone will SEW you if anything goes wrong...&quot;

A real stitch-up eh? Har har.

Edited by - Dippy on 12/07/2006 09:54:59

Thanks for all your input. Some additional information for your peace of mind regarding the design and operation of my &#8216;mower&#8217;.

* Mercury tilt switch fitted on main power line from battery to &#8216;Kill the Beast&#8217; if it is picked up.
* Keypad to input run code to stop unauthorized use.
* Drive motors are run by the Devantech MD23 which require a move command every 2 seconds otherer wise it will stop. Therefore is electronics fail the maximum time it can continue to run is 2 seconds.
* Using the RF loop for boundary marking also leaves a RF signal over the garden while the transmitter is powered. So by having a third receiver coil continually checking for a signal will allow the failure of the transmitter to be detected and therefore initiate the &#8216;STOP&#8217; routine.
* Bumpers on the front &#8211; bit like whiskers will allow it to bump and reverse away from obstacles.

I have not seriously considered the use of a speech synthesizer to give a warning but there are options to have an ultra-sonic surveillance zone around the mower to trigger an audible warning.

My interest stems from my Mothers robot mower, MOWBOT. <A href='http://www.mowbot.co.uk' Target=_Blank>External Web Link</a>
This does all the things I am trying to recreate but for less than the GBP 1,300 she paid for hers. The buried wire fence if the bit that I am in need of help with. I have looked at the site mentioned above and read their diary but they have not answered my request for help in understanding how their sensors work.

I found this last night - <A href='http://www.philohome.com/sensors/filoguide.htm' Target=_Blank>External Web Link</a>
I have seen even simpler designs but was unable
to locate any at the moment.
Aside from the logistics and safety issues
this method does not have to be complicated.
A relatively simple set of TX RX circuits
ahould only cost a few dollars each.
I will keep looking -

Just a couple more comments on the failsafes, not to get at you particularly, but to highlight to others who may try and follow in your path with less experience ...

<i>Drive motors are run by the Devantech MD23 which require a move command every 2 seconds otherer wise it will stop. Therefore is electronics fail the maximum time it can continue to run is 2 seconds. </i>

Unless the failure mode is such that the electronics ends in a condition where it is forever sending &quot;Go&quot; to the motors.

<i>Using the RF loop for boundary marking also leaves a RF signal over the garden while the transmitter is powered. So by having a third receiver coil continually checking for a signal will allow the failure of the transmitter to be detected and therefore initiate the &#8216;STOP&#8217; routine. </i>

Except where the failure mode is that the signal present detector reports it always is regardless.

The traditional approach to solving the problems of whether the electronics are 'telling the truth' or not is to implement that part of the functionality three or more times over ( preferably with different software and with different hardware designs ) with a majority vote being taken to determine what reality really is.

One soon starts to see why the cost of some products looks much higher than one would first expect. The need for such extreme failsafes is proportional to the damage which can be done if something does go wrong, balanced against the risk/likelihood of something going wrong.

Sorry if I do at times sound like a whinging old man, but it's caution which has to be expressed in a public forum and isn't directed at anyone personally. Once people understand the risks and what's involved and understand how to look properly at assessing such things, barring legislative rules, it's up to them to take responsibility for what they do. &quot;Look before you leap&quot;, is really the underlying message; knowing how to look and what to look for is the hard part. On the opposing hand is, &quot;if we never took any risks, we'd never do anything&quot;. Getting the balance right is the goal.

You're quite right to point out the dangers and all that. I totally agree.

The designer should think hugely carefully about fail-safe scenarios - even if it has to include non-electronic 'old fashioned' safety cut-outs.
And of course the Big Boys have more money/time/experience to sort it all out. Plus the fact (hopefully) the product conforms to international safety standards etc etc.

I'm sure if I tried hard enough I could chop my fingers off using the pukka Mowbot.

But, apart from throwing caveats at all designers, lets not put him/her off too much eh? We can always visit them in A&amp;E and say &quot;I told you so!&quot;.

Indeed, the home designer, may actually think of something that the Big Boys haven't.

All I'll say is be careful, think and check everything and don't penny pinch and don't bodge! Start small and develop it, don't go head-first into a big project. And start with something without rotating blades!!

Start by making a bot which sense the wires.

Sadly, you can have all the safety in the world but if the end user is a complete pratt then...

I understand your consern about safety, just as i do - if it cant be done safely then dont do it. the blades and blade motor are going to be the last part of the project, once the platform has been put through it's paces and all safety systems checked.

As you say, it is one step at a time.
1. get the platform to move.
2. add bumpers to platform to bump-reverse-turn.
3. add burried wire sensors.
4. add Transmitter check sensor.
5. vigourouslt test in final enviroment.
6. Add cutting motor.
7. Vigourously test full assmsbly with no cutting blades/wire/cord.
8. add blades.

Big red cut-off button on top and option of an IR receiver from remote control to kill power

It sounds good and an interesting project, so good luck with it. One trick in the early days of testing ( I used it on a much smaller robot ) is a phono plug / jack or similar which can be pulled out by a bit of string to kill power if it races off into the distance. Sod's Law dictates that with the first on-the-ground test it will reverse straight into your shins at speed - if it doesn't, suspect there's a programming bug

<BLOCKQUOTE><font size=1 face=arial>quote:<hr height=1 noshade>...is a phono plug / jack or similar which can be pulled out by a bit of string to kill power <hr height=1 noshade></BLOCKQUOTE></font><font face='Verdana, Arial, Helvetica' size=2>

Unless the string breaks!!!

That's a good point Rick.

Hippy didn't include a breaking strain in his Risk Analysis.
And of course if you pulled it out a bit too sharply it might come back and hit you in the eye!

Oh give up the project - its just too dangerous!

(A JOKE by the way)

Also when the thing is heading at you at speed it's hard to reel in the string fast enough to take up slack and get pull

Stay safe guys... [model aircraft heading your way are the most comon...]

I did think of breaking string, but thought I might sound a little too anal and fun-spoiling if I mentioned it

It's a fine example of knowing where or choosing where to draw the line, but it does highlight that there is always something else which could go wrong. Brainstorming disasters is always good fun, but sometimes the most seemingly ludicrous scenarios ( what if the batteries simply exploded ? ) do have some merit to them ( ... such as when a mobile phone is dropped ... or the user has used third-party batteries ).

Your risk analysis isn't complete.

Even if you built a &quot;dangerous&quot; mower-- would it be more dangerous than pushing around the old clunker in your flip-flops?

Sorta like the issue of nuclear power plants emitting radioactivity and ignoring the radioactive biproducts of coal fire plants.


(there- now I've started something!)

To my ammazement we had one of a set of Alkaline batteries explode reciently - A first for me and not previously considered as part of my risk assessments!! Goes to show.

bttery company currently investigating circumstances.

But have you admitted to the battery company that you were trying to weld two batteries together using oxy-acetylene when it happened?

Michael, the lego circuit looks good. i will add it to my reading for this weekend.

Dippy - Of xcourse not - Actually the batteries (a set of 4) were connected to a 6 volt MES lamp. As the lamp wasnt lighting I guess there was a short although we found nothing after the event. (this the company know)

We also intentionally short circuited a new set of batteries for several minuits - They get HOT - VERY HOT but no split in the case.

I expect it was a rogue case and just one of those things - But it can happen and we are now aware of the dangers.

Try putting one of the batteries in the wrong way around and then short the pack. That is almost gauranteed to make it explode!

Don't encourage me,,!

<i>Michael, the lego circuit looks good. i will add it to my reading for this weekend. </i>

There's an error in the generator schematic. Pin6 (threshold) is connected to pin7 (discharge). Pin6 should be connected to pin2 (trigger). The circuit in the simulation link is correct, though.

The 'no-threshold rectifier' in the sensor schematic is interesting.

The manufacturer say they shouldnt explode because they ar fitted with a vent to release high pressure. They say!!!

So was the Titanic.

a blob of solder over the vent works nicely :-0