Radio/magnetic field: detect direction of antenna

[petterg]

To follow a wire you need some phase reference to identify which side you are on. So consider two small pickup coils. One is mounted on the vertical axial plane of the vehicle, the other is horizontal. The vertical plane coil detects a phase reference signal from the wire.

Wow! Great!
Does this require the horizontal coil to face the same direction all the time? In this case it's not going to do that. Currently the abs(X1) is found by two horizontal coils, separated by 90 degrees. Then they are added together (digitally) as Pythagoras. This give a value that is independent of which way the coils are facing, but it's also loosing the sign component of the signal.

If I understand your description, the method makes use comparing phase in the horizontal and the vertical coils, and that will require the horizontal coil to face the same direction at all times. I hope I'm wrong.

All though, comparing phases from all three coils might be the way to go?

A momentary current pulse, say once each second might be detected directly by the vertical coil, the polarity of the detected spike would identify direction along the wire.

That would be the sawtooth generated signal, right?
Cause, a momentary current pulse in the wire will have just as much positive as negative change of magnetic field, hence require the detector to see what came first. While the sawtooth in the wire makes the magnetic field change fast in one direction and slow in the other. (I'm not sure about this at all.)

 

[Baluncore]

The phase of the signal is determined by the direction the flux passes through the coil, use your “Right Hand Rule”. The horizontal plane coil, flat with the floor, will always be above the wire so it will have a phase and magnitude determined by the left/right horizontal position error. The vertical plane coil will have a phase determined by the direction faced along the wire. That phase reverses if you turn the vehicle 180°, but so does the left/right phase picked up by the horizontal plane coil, so the phase detection still steers correctly along the line.

No sawtooth wave is needed, just the DC pulse. The AC coupling of the mixer will fix the asymmetry and be ignored by the mixer. By following and holding peaks of both polarity from the vertical plane coil, their comparison will show which direction the vehicle is facing.

A perpendicular pair of coils at 45° to the horizontal could be used. They would need to be converted to sum and difference by a couple of op-amps. The low-pass filtered product of the sum and difference voltages would be the left/right steering signal. This is really just an electrical rotation of the coil orientation. Phase detection is identical.

 

[petterg]

I don't see how the phase on the vertical coil will change when turning the vehicle 180 degrees in the horizontal plane. Using the right hand rule with my thumb up, my fingers point to the left no matter how much I rotate my body in the horizontal plane. As far as I can see the phase in horizontal coils change when the vehicle turn, while the phase in the vertical does not. And that makes this a little more complicated.

When turning the horizontal coil the signal will become weaker as the vehicle turns, until it's 0 when vehicle has turned 90 degrees, then it becomes stronger with the inverted phase when turning towards 180 degrees. To avoid this dead signal at 90 degrees I have an extra horizontal coil that is placed perpendicular to the other horizontal coil. As the signal weakens on one, it's getting stronger on the other, and using Pythagoras we know that I(h1)^2 + I(h2)^2 = constant at a position independent of the vehicles direction.

Maybe multiplying all three coils will be the solution?

 

[petterg]

Here's why horizontal + vertical coil won't work, even 2 horizontal + 1 vertical won't work:
The variables affecting the detecting coils are:
- signal phase, 1D, positive or negative
- vehicle position in X-direction, 1D, positive or negative
- vehicle heading, 2D, positive or negative values in both X and Y direction

This sums up to 4 dimensions. To describe an unique position in this environment there is a need for a 4 bit value. Hence there is a need for detecting the signal i 4 different ways. 3 coils give 3 a 3 bit value. We need one more!

I think that 4th sensor has to be an unsymmetrical wire signal. Any opinions on the sawtooth idea?

 

[Baluncore]

Using the right hand rule with my thumb up, my fingers point to the left no matter how much I rotate my body in the horizontal plane.

Your thumb represents the current, your fingers the magnetic field. The wire is along the ground surface so your thumb should point away or towards you. The magnetic field from the wire rises on one side and falls on the other.

Maybe there is a misunderstanding here of the orientation of coils.

As far as I can see the phase in horizontal coils change when the vehicle turn, while the phase in the vertical does not. And that makes this a little more complicated.

A horizontal plane coil lies flat on the ground surface with a vertical axis and is totally independent of heading direction. If the horizontal plane coil is on one side of the wire it will have the opposite phase to the other side of the wire. There will be no signal or phase when centred over the wire.
A vertical plane coil is mounted on the lengthways vertical plane of the vehicle with it's axis aligned left/right. It has a phase determined by the direction the vehicle is pointing. It becomes the phase reference on the vehicle.

 

[petterg]

Maybe there is a misunderstanding here of the orientation of coils.

I guess it is. When I say vertical coil I think of the coil produced when holding my finger vertical and spin a wire around it. The wire then will be somewhere near the horizontal plane, while the axis and magnetic field in the coil will be vertical. I guess you call this coil horizontal because it's wire is close to horizontal, while I say it's vertical because of the direction of it's axis.

Still, the method you suggest requires a vertical and a horizontal coil to change phase simultaneously, which they won't.

The phase of the signal is determined by the direction the flux passes through the coil, use your “Right Hand Rule”. The horizontal plane coil, flat with the floor, will always be above the wire so it will have a phase and magnitude determined by the left/right horizontal position error. The vertical plane coil will have a phase determined by the direction faced along the wire. That phase reverses if you turn the vehicle 180°, but so does the left/right phase picked up by the horizontal plane coil, so the phase detection still steers correctly along the line.

With that misunderstanding cleared out, let's check if there is another misunderstanding too. As far as I understand the second statement in bold from the quote above says the exact opposite of the first statement in bold. They both can't be correct. To my knowledge the first statement is correct, the second is not. If they both were correct the problem would be solved. (I hope you see a second misunderstanding here.)

 

[sophiecentaur]

Is it absolutely essential to use just one wire? If you can use some 2 dimensional array of wires, the problem becomes much more tractable because you can be dealing with time / phase rather than the amplitude of a signal. Hyperbolic navigation systems work so well that they were used very successfully until GPS came along. The implementation becomes almost trivial with the possibility of a bit of intelligence on board (Arduino board, for instance). You no longer need to be dealing with 'direction', which relies on complicated transmitting or receiving antennae and you just rely on 'timing', which can have a relative precision of less than 1ns (giving 1ft)

 

[petterg]

Radiobased hyperbolic navigation would surely be the best. But to my knowledge this would require free visual to transmiters That will not be the case here. It has to work even when obstacles move around. Making a grid if wires is not an option either.

I still haven't found any reason why the sawtooth-signal won't work. That is a good sign.

 

[sophiecentaur]

Why would it need "visual"? The presence of any conducting object would have more perturbing effect on the level of received signal than on the phase, so the time method still works out better. You would not need a 'grid' of wires - just three sources (minimum) or four (better) sources - short monopoles or vertical axis coils, placed one each end and one on either side.

 

[petterg]

You're right. Hyperbolic will be a lot better. If it works I can kick out all other attempts to keep track of position. They are all estimates with a huge error.

All I know about hyperbolic is: A antenna receives a wave from three (or more) sources and calculate the distance to each of them.
My biggest concern is that the timer in an ardurino won't have the resolution required to do the timing.

How does the receiver know which wave came from which sender? Do all senders send on different frequencies? Do the senders communicate in any way to synchronize when they send out the waves?
Is there a minimum/maximum distance sender-sender or sender-receiver?

 

[sophiecentaur]

The arduino needn't do the timing. The Arduino only needs to get the phase information from some other circuit (a mixed down version of the RF). In the Decca Navigator system (which used low frequencies appropriate for the large distances involved), four different carrier frequencies were used and the phase between them and four references was measured by having four receivers. In the Decca system, one transmitter was the master and the others synced with it. In this case, all four could originate from the same master oscillator and could have nice integer relationships. There's a lot of basic stuff about hyperbolic systems and the Decca Navigator on Wiki.

 

[petterg]

Now I've read quite a bit of several versions of hyperbolic navigation. It comes down to the fact that arduino runs at 16MHz, We must take into consideration that a signal may be received just after one clockcycle. Then it won't be processed before the next clock. If this should be based on timing then the time between two clocks alone makes an error of 18,73m. That's worse that what I already got without knowing which side of the wire the robot is on. Either I'm to dumb for understanding this, or timing based navigation cannot be used.

The other option is based on interfering waves. To get a reasonable resolution we're foreced to run on quite high frequencies. We hit frequencies that cannot be freely used. So lowest possible frequency that can be used is 2,4GHz band. With this high frequency we get pretty small grids of waves. Seems like some old wlan routers will be useful. But I can't figure out how to decide which point of intersecting waves the sensor is located in. It seems to me that when this kind of navigation was used, ppl knew approximately where they were located, so when they hit a point where waves were intersecting they know from their manual approximation which intersection that was.


I think I'm falling back to the sawtooth again. Still nobody have said anything of why it shouldn't work.

 

[sophiecentaur]

You are confusing phase resolution with processor speed. Unless your robot is moving very fast (a large fraction of a wavelength each processor cycle) then the processor speed really doesn't matter. It just looks at the phase of a much lower frequency signal (beaten down version). Time difference, appears as phase of the IF frequency, which can be chosen to be anything convenient (it's a Delta t 'magnification' process' due to the frequency ratio). This is the same idea as when a sampling scope looks at extremely high frequencies and sub-samples at a manageable frequency for its circuitry. A time domain reflectometer can resolve distances of a fraction of a cm in this way, too. I think the processor speeds used to be just a few MHz in the TDRs I used in the 80s.
For the powers you are interested in (very short range operation), there is a pretty wide choice of operating frequency - the WAN frequencies would seem a good choice as the antenna would be pretty small.
BTW, you needed to know nothing extra about where you were for Decca to tell you - as long as you were on the appropriate map of the North Sea / English Channel etc. The channels of the transmitting stations would tell you that, in any case.
But Decca was real steam radio stuff, operating with carriers less than 1MHz and with ancient valve equipment, originally. It also had virtually no built in 'intelligence'.

Your suggested amplitude-based method would be full of sources of error, due to parasitic components around the set up and unknown directivity patterns. I don't know of any examples of a system for assessing position which relies on signal strength measurement. I must say, I haven't looked in detail at your idea because it doesn't, afaik, have any precedent. If you can quote one then you might be in with a chance, though.

 

[Baluncore]

(I hope you see a second misunderstanding here.)

Do you agree with the following four statements.

1. Consider a vehicle with a coil lying in the horizontal plane, having a vertical axis, near the signal wire. The heading of the vehicle is not important. The phase and amplitude of the induced voltage will depend only on the axis position relative to the transmitter wire. When the vehicle/coil is on one side of the wire the phase will be the opposite to when on the other.

2. Consider a coil mounted on the vertical plane of the vehicle with the coil axis left/right. The induced voltage will have a phase determined by the heading of the vehicle along the wire.

3. When following the wire the left/right steering information derived from the vertical axis coil will depend on direction of travel. If the left side magnetic field is up then the right will be down. Travelling in the opposite direction will view the right side as up and the left side as down. That reversal must be handled or the vehicle will not follow the line.

4. The product of the voltage induced in the two coils performs that correction automatically.

 

[Baluncore]

I don't know of any examples of a system for assessing position which relies on signal strength measurement.

The Lorenz beam landing system was used, (with cross beams), for bombing navigation over Britain in WW2. It compared the amplitude of two signals to define a straight line that was accurate to 300m from over 100km away.

I'm trying to design a circuit (for use in a robot) that can detect if a wire carrying an AC-current is to the left or right of the detecting circuit.

The OP here specified following a wire, not deriving a 2 dimensional navigation system to rival differential GPS.

Following a wire is trivial. It is done by telecommunications engineers who need to follow and mark the position of buried cables.

 

[petterg]

You are confusing phase resolution with processor speed. Unless your robot is moving very fast (a large fraction of a wavelength each processor cycle) then the processor speed really doesn't matter. It just looks at the phase of a much lower frequency signal (beaten down version). Time difference, appears as phase of the IF frequency, which can be chosen to be anything convenient (it's a Delta t 'magnification' process' due to the frequency ratio). This is the same idea as when a sampling scope looks at extremely high frequencies and sub-samples at a manageable frequency for its circuitry. A time domain reflectometer can resolve distances of a fraction of a cm in this way, too. I think the processor speeds used to be just a few MHz in the TDRs I used in the 80s.
For the powers you are interested in (very short range operation), there is a pretty wide choice of operating frequency - the WAN frequencies would seem a good choice as the antenna would be pretty small.
BTW, you needed to know nothing extra about where you were for Decca to tell you - as long as you were on the appropriate map of the North Sea / English Channel etc. The channels of the transmitting stations would tell you that, in any case.
But Decca was real steam radio stuff, operating with carriers less than 1MHz and with ancient valve equipment, originally. It also had virtually no built in 'intelligence'.

The robot moves about 30cm/s at top speed. (At least that's what I hope. The motors haven't arrived yet.) Speed shouldn't be an issue.

When you're writing it sound like you have the perfect solution. But when I read it all seems far to complicated to implement. Where do I start? There seems to be a huge load of ways to do this. I can't even figure out which one to concentrate on. Not that I need to follow any standards.
Goal is to make the robot navigate in a area of 30x30m. An error < 1m would be very good.

Your suggested amplitude-based method would be full of sources of error

Then I rule out amplitude-based.

I must say, I haven't looked in detail at your idea because it doesn't, afaik, have any precedent. If you can quote one then you might be in with a chance, though.

What does that mean? I lack some language skills to understand the meaning. (And google translate didn't help this time.)

 

[petterg]

The OP here specified following a wire, not deriving a 2 dimensional navigation system to rival differential GPS.

A differential GPS would definately be a great solution for this problem. All though, I think it will be too complicated. Something that can make the robot know exactly where it is, is my dream.

Do you agree with the following four statements.

1. Consider a vehicle with a coil lying in the horizontal plane, having a vertical axis, near the signal wire. The heading of the vehicle is not important. The phase and amplitude of the induced voltage will depend only on the axis position relative to the transmitter wire. When the vehicle/coil is on one side of the wire the phase will be the opposite to when on the other.

2. Consider a coil mounted on the vertical plane of the vehicle with the coil axis left/right. The induced voltage will have a phase determined by the heading of the vehicle along the wire.

3. When following the wire the left/right steering information derived from the vertical axis coil will depend on direction of travel. If the left side magnetic field is up then the right will be down. Travelling in the opposite direction will view the right side as up and the left side as down. That reversal must be handled or the vehicle will not follow the line.

4. The product of the voltage induced in the two coils performs that correction automatically.

1: yes
2: yes
3: yes, for the special case when the sensor is rigth above the wire. "No" when it's a litle bit on the side.
4: yes, when #3. While #3 is not true, no.

In #1 you say " The phase and amplitude of the induced voltage will depend only on the axis position relative to the transmitter wire."
If #3 should be true for other locations than rigth above the wire, then that quoted part of #1 cannot be true. #1 is true, hence #3 can not be.

So I guess the missunderstanding here is that you think the robot will be on the wire, while my question is how to detect on which side of the wire the robot is.

Here is an example of location and the information the robot can detect with my current knowledge:
Say you have a wire going north-south, previously called the y-axis. The wire is 3mm thick. The robot is 600mm long, 400mm wide. The distancecalculation indicates the robot is between 6000mm and 8000mm from the cable. (This distance calculation is subject for disturbtion.)
The vertical axis coil indicates magnetic field going up.
The horizontal axis coil indicates a weak magnetic field towards the front of the robot.

What I'm not able to figure out: Is the robot on the east or west side of the wire?

Well, if the current in the wire is pointing to the north, the magnetic field will be up on the west side. If the current is heading south, the magnetic field will be up on the east side. So with the robots current knowledge there is two posible locations.
If the wire signal is of a kind so that the robot can identify its phase, then it would know it's location.

The best way so far, within the limits of what I think I'm able to implement seems to be the sawtooth-signal. Although I think Sophiecentaur might put me on another track.

 

[Baluncore]

So I guess the missunderstanding here is that you think the robot will be on the wire, while my question is how to detect on which side of the wire the robot is.

I thought it would be close to the wire and would be attempting to follow the wire. My suggested arrangement detects which side of the wire the sensor is on. If it is exactly on the wire the output will average zero and the question “which side” will be answered as 0 = neither.

I'm trying to design a circuit (for use in a robot) that can detect if a wire carrying an AC-current is to the left or right of the detecting circuit.

I was assuming that the vehicle would be moving parallel with or following above the wire. I saw no precision ranging requirement for the distance from the wire, certainly not of the order of 5m. A Lorenz type system could be integrated into the system later that would achieve that; http://en.wikipedia.org/wiki/Lorenz_beam it would be a different system but a similar concept.

Now let's add a minor complexity to the simple wire. When current is flowing one way a small, (10mW), VHF radio transmitter transmits a carrier. When the current flows the other way there is no carrier transmitted. This gives you an independent phase reference so a vertical axis coil can now tell which side of the line it is on.

 

[jim hardy]

Perhaps it's simpler.

How far away is the wire?
I assume it's a single conductor not a two conductor lampcord where the resultant magnetic field would be nearly zero...

Were the robot pretty well non-magnetic , two identical magnetic sensors on opposite sides of the robot would receive signals differing in strength by square of their respective distances from the wire, which would differ by width of robot.
A simple amplitude comparison might do the trick if the distances are not great compared to size of robot..
Sum the two signals and square up the result with a Schmitt trigger, use that for CLOCK input to a D flip-flop
Difference the two signals and square up the result with a Schmitt trigger, use that for DATA input to same flip-flop.

OUTPUT of flip-flop is direction L or R.

I think that's sort of like the multiplication suggested by Baluncore

but a TV show on hammerhead sharks made me think of it.

Might need to delay the summed signal by a fraction of a line cycle so that comparison is made well after sinewave's zero crossing.

A job for LM324's and 555's . (I know , I'm showing my age ).

Is this practical or am I having a midsummer night's dream ? old jim

 

[petterg]

Here's a document that gives a good overview of various ways of using wifi for navigation
http://www.snet.tu-berlin.de/filead...12/snet-project/wifi-positioning_henniges.pdf

 

[petterg]

Thank you, Jim

I think your onto something.
I need to look into the circuit you describe and see if I understand it.

It is a single conductor wire. Well it can be anything. My intetion was to use a single conductor. If it turns out that an other wire makes stuff easier I'll change it.

While being a single conductor it also has to be a loop. Hence a coil. Being a coil, the magnetic field can be made stronger when using a multi-conductor and connect the conductors in series. So maybe a single conductor is not prefered.

The robot will be at most 20m from the wire. That is 30 times the size of the robot. I guess there needs to be at lest 3 sensors in order to make the theory work while the robot is able to turn.

 

[sophiecentaur]

The robot moves about 30cm/s at top speed. (At least that's what I hope. The motors haven't arrived yet.) Speed shouldn't be an issue.

When you're writing it sound like you have the perfect solution. But when I read it all seems far to complicated to implement. Where do I start? There seems to be a huge load of ways to do this. I can't even figure out which one to concentrate on. Not that I need to follow any standards.
Goal is to make the robot navigate in a area of 30x30m. An error < 1m would be very good.

Then I rule out amplitude-based.

What does that mean? I lack some language skills to understand the meaning. (And google translate didn't help this time.)

You are right; my proposal is certainly complicated and it would do more than just to tell you which side of wire your robot is - which may not be needed.
If all you really want to know is that then the solution may be simpler - because you do not need to measure distances accurately.
The problem still remains that the robot needs to know something of its orientation. The idea of an asymmetrical waveform for the current on the wire is a good one because it gives information about the sign of the field vector. A single coil will not tell you which way to go to get to the wire (it will just tell you which side you are on) and your robot would need to 'hunt' in order to find the slope of the amplitude / distance relationship. Three coils in quadrature could tell you the direction of the magnetic field and give you a rough idea of its strength - hence the distance from the wire plus which side you are on. However, if your robot is in or near the same horizontal plane as the wire then the field will appear vertically up or vertically down, which gives no information about the direction to go to reach the wire. If the robot carries the coils high enough to give an 'angle' then - less of a problem.
Imo, there is no substitute for some form of phase /timing measurement, if you want to know your distance from the wire. A single measurement of distance from the wire, over a limited range, could be a compromise.
Take two oscillators, using any high frequency you choose, with their frequencies differing by exactly 100MHz (say) and feed them to the wire. The distance of the robot from the wire can be found by the phase difference, which would repeat every 3m. Again.there would be a non-trivial bit of circuit design (quite do-able, though).

The problem would become much more tractable is you wanted to keep the robot between a pair of parallel wires, using a phase difference approach.

The final paragraph in my other post was just saying that, if you can't find evidence of anyone else using just amplitude, then you are unlikely to make it work. I am ignoring the systems used for automatic aircraft landing because they involve a complex pair of transmitting antennae to form the right beam shape.

 

[jim hardy]

yes, the current must get back from whence it came even if by multiple paths...
Maybe you can get creative with the solid geometry of the room.
It's time for some experiments, I think.

Baluncore and Sophie are better versed in math and field theory than I, they probably are fluent in "Poynting vectors" and "intensity" .

On a low level practical note - What are your constraints?

Were the 'wire' to be a light rope made of LED's with something different about the ones pointing left and the ones pointing right, it'd become a visual sensing task .

Issue to solve: The current in the wire has to be AC.

Is that a constraint? Fluxgate or Hall magnetometer can sense a DC field, three of them can resolve its direction.

Good luck and have fun !


old jim

 

[sophiecentaur]

yes, the current must get back from whence it came even if by multiple paths...
Maybe you can get creative with the solid geometry of the room.
It's time for some experiments, I think.

Baluncore and Sophie are better versed in math and field theory than I, they probably are fluent in "Poynting vectors" and "intensity" .

On a low level practical note - What are your constraints?

Were the 'wire' to be a light rope made of LED's with something different about the ones pointing left and the ones pointing right, it'd become a visual sensing task .


Is that a constraint? Fluxgate or Hall magnetometer can sense a DC field, three of them can resolve its direction.

Good luck and have fun !


old jim

I think the message is that 'on the whole, we'd rather not start from here'. I reckon that what's called for is a modification of the initial requirements. There are so many more ways of approaching this than insisting on using a single wire. As Jim says - you also need a return current path, which is highly relevant in a practical situation. You can still have something that looks like a wire but why can't it be more complicated? Two wires and a third wire between them to act as a reflector could be arranged to fire two beams in the left and right directions but this, of course, would involve some RF and the spacing of the wires and reflector would need to be a quarter wavelength - but why not? 3cm microwave systems are available as School demo kits (Gunn diode) and would be suitable for a few experiments. Then there's the optical (infra red) beams solution, which would actually be a piece of cake.
Basically anything but the single wire.

 

[petterg]

...

The reason I started out this thread with the purpose just looking for which side of the wire the robot is on was that I already have (in theory) methodes to estimate the location. Those methodes are:
+ Estimate distance from wire based on received signal amplitude, probably 40% error.

+ On occations when none of the engines are running, get a compass reading. A compass in front and back can hopefully averange out distrbance from souroundings.

+ Count wheel rotations. That will give robots rotation and traveled distance in each direction.

+ GPS readings - look at difference with previuos reading. GPS reading can be 100m off. But it doesn't really matter if it's 100m off as long as it's still 100m off afer moving 1m. Sometimes the GPS position can jump 40m in a second. Such events has to be filtered out. I considered multiple GPS'es, but after walking around with three GPS's I realized that they tend to get the same error. However, two GPSes 60cm apart may have the effect that only one of them are located at a spot disturbed by an obsticals.

+ Every time the wire is crossed, calculate each of the aboves errors. Count for thees error when moving on.

= I'm hoping the sum of all of this will return a position that is accurate down to 50cm. It's acceptable that location is off by a few meters as a startup. This should be adjusted for when the wire is crossed. Then the presission will drift until the next time the wire is crossed.

IF I manage to see how to implement radio navigation with better pressision I can throw away all the above. Hence the logic might be simplified by doing something complex. Alternatives to location implementations are very welcome!

When I started the thread I was convinced that radio navigation would be far to complicated to implement (When I slightly understand the basic theori, I'm still FAR away from being able to make an implementation.) Also I though it would be less accurate than the sum of the above methods.

The idea of using dual frequency is also a good one. (I understand the basic theory without looking it up.)
The field will be close to vertical most of the time. Sensor at most 0.3m above horizontal wire plane, and up to 20m away. But who says there can't be several coils at different angels on the robot?

I've ordered quite a bit of components today so I can start playing in a few weeks. (Ships from Hong Kong)

 

[petterg]

..

The wire will be digged down a few cm in the ground. It won't be visible.

The constraints is that this needs to not be distrubet by the wires to the street lights. The equipment should be as discret as posible (wires across the area will block for people moving around). Digging down a wire or two is ok. Digging down a grid of wires is not ok.
And there's quite a bit of wlans in the area that could be used for positioning. And finally, it shouldn't send out signals that's disturbing the suroundings.
Other than that limitations can be adjusted.

Hall sensors (or anything that uses DC) will be distrubed by random magnetic fields.

 

[jim hardy]

I had no idea what you were up to - maybe robot wars or something.

I was even wondering if we could detect the Earth's magnetic field in the vicinity, and modulate it slowly with our loop. That'd give us robot heading information twice per cycle at zero crossings of our modulation... just daydreaming...

Sounds almost like you're taking the wireless dog fence to next stage... I wonder what's to be gleaned from taking one of those apart... it might beat reinventing the wheel.

 

[sophiecentaur]

You now say that you are considering using other positional information. That will give you a much better chance. Dead reckoning (wheel rotations plus direction) would give you some good results,
This should be a really fun project, I think.

 

[Baluncore]

There are too many possible solutions for this undefined problem. Random guesses at moving goal posts should be quite successful at preventing completion of the project.

Without a specification of the performance requirements and the operational environment it will not be possible to advise a sensible solution.

Are we permitted to know what the vehicle is for?
Is it armed and dangerous?
Is it reasonable for us to assume it is a robot vacuum cleaner or lawn mower?
What are the maximum dimensions of the field?
Is knowing position sufficient or is vehicle orientation also needed?
How accurately must position and orientation be known?
How fast does the vehicle travel?
Is it inside a building or outside?
Must it work in the dark and in bright sunlight, night, day or both?
What obstructions to visibility are there in the field?
Can there be several points in the area, three of which are always visible from the vehicle?
Will the vehicle be alone or with others, friends or enemies?

 

[sophiecentaur]

@petterg
If you want this thing to operate 20m away from your wire you will need hundreds of amps of current in order to exceed the Earth's magnetic field. (use this link to calculate it) You are forced to use RF techniques at this distance.
I can echo Baluncore's frustration at the lack of specific details you are 'leaking out', when the solution to sort of project relies totally on the values of the variables involved. Just buying a load of components is not the way forward; planning is necessary.