Quickly
3 arduino's in a single file. A, B and C
Can B tell if A is to its front/back or to it's left/right?
What positioning shield would be able to detect and transmit this data in a noisy area (Infared wouldn't be an option here)
Thanks,
Sovrac
Can B tell if A is to its front/back or to it's left/right?
No this is not easy, this is a very difficult project.
It depends on the range you need and how far an angle the behind / forward is.
Infared wouldn't be an option here)
Ultraviolet?
Grumpy_Mike:
Can B tell if A is to its front/back or to it's left/right?
No this is not easy, this is a very difficult project.
It depends on the range you need and how far an angle the behind / forward is.
The range would be a radius of up to 10 feet.
What would it take to accomplish this?
The goal would be to detect other arduino's withing its range and the location of that arduino (to it's left/right).
AWOL:
Infared wouldn't be an option here)
Ultraviolet?
From my understanding the UV signals could be easily interrupted just like IR, do you have a UV solution in mind that you could share?
The range would be a radius of up to 10 feet.
There have been quite a few people asking for this sort of range sensor but as far as I know there is no solution.
Grumpy_Mike:
The range would be a radius of up to 10 feet.
There have been quite a few people asking for this sort of range sensor but as far as I know there is no solution.
.
Is there currently a solution for a radius of 1, 2, 3 or 5 Feet?
I'm dreaming here, but... I always wanted to make a positioning device using an old style circular antenna as a radiogoniometer. Mounting the antenna on top of the servo or stepper motor would make a robotized location device to a beacon mounted in another device.
I don't have enough RF knowledge to make that, but having something like this would be pretty sweet.
By the way, back in time most of the sci-fi robots had a round antenna in the head 8)
With a couple of antennas, if you just need low accuracy, you might solve it with RFID tags.
pgmartin:
I'm dreaming here, but... I always wanted to make a positioning device using an old style circular antenna as a radiogoniometer. Mounting the antenna on top of the servo or stepper motor would make a robotized location device to a beacon mounted in another device.
I don't have enough RF knowledge to make that, but having something like this would be pretty sweet.
By the way, back in time most of the sci-fi robots had a round antenna in the head 8)With a couple of antennas, if you just need low accuracy, you might solve it with RFID tags.
Low accuracy, meaning it would provide innacurate responses the majority of the time... I dont see the use of that, if the purpose is accurate position detection...
Sovrac:
You mentioned in front/back, left/right. That sounded like you don't need minute of angle accuracy, right?
What kind of accuracy do you need?
If IR or any light solution is not viable, is RF? What do you mean with noisy environment? Is it full of physical interference between the Arduinos?
Is this a robot swarm project?
if the purpose is accurate position detection
First mention of that.
How about some long, stiff rods, attached to potentiometer shafts?
pgmartin:
Sovrac:
You mentioned in front/back, left/right. That sounded like you don't need minute of angle accuracy, right?
What kind of accuracy do you need?
If IR or any light solution is not viable, is RF? What do you mean with noisy environment? Is it full of physical interference between the Arduinos?
Is this a robot swarm project?
The devices will be moving in a random direction from each other, with physical interferences between them.
When one has made a link with another (an arduino has enter the others range), they would have to know what direction the other one is in (to its left/right)
MOA could be overkill, would an RF transmitter coupled with another device provide a solution?
Is there currently a solution for a radius of 1, 2, 3 or 5 Feet?
No but if you are down to the 3 to 6 inches you could probably use induction loop proximity detection.
would an RF transmitter coupled with another device provide a solution?
I can't see how.
How about an overhead camera with fiducials on the tracked objects?
Or the inverse, like the BBC developed, with upward pointing cameras, and fiducials on the ceiling?
Of course, this is way beyond an Arduino now, but I suspect this problem always was.
I suppose the devices are moving on the floor. Can you install something in the floor?
What does it mean entering the other arduino's range? Is this a distance, that might include obstacles between them?
I think this is an anteresting problem, and maths will provide an interesting answer...
Lets look at GPS first. A bunch of sattelites sit above earth with a clock on them. One sattelite will then transmit the time. Your GPS receiver compares the time on its clock to the time the signal says. Knowing the speed of light, we can work out how far the sattelite is. Other sattelites transmit their time, your receiver builds up a list of distances, then uses triangulation or trialteration to work out your location. Now on to the arduino...
Each arduino is equiped with an RTC, ultrasonic transmitter and an ultrasonic receiver. Device A will transmit the time on its RTC, followed by its identity (A). The other devices will work out the distance away it is, based upon the time their RTC says when the signal is received, and the time the signal says. Each other device will do the same. Now each device knows how far away the other two are, and we can get something like this:
Then a miracle occurs, and you know the angle each device is relative to straight ahead. (Note that the miracle will be some trigonometery. I do not know how to spell it, never mind how to work something out with it. Ask someone else 
). (Also note that the ultrasonic signal might be reflected from a wall. To prevent receiving the same signal several times and getting confusing results, allocate each device a certain time to transmit and ignore any data received out of this time. This is time division multiplexing). Problem solved.
Onions.
Could you use ultrasonics, the range is about right I think.
As there are only 3 Arduinos each could ping out a unique code.
By listening an Arduino might be able to figure out the position of the other two units.
I am not sure how far apart its "ears" would need to be.
Perhaps if you had four listening sensors that were fairly directional the volume, rather than time delay, would let you the rough location.
pgmartin:
I'm dreaming here, but... I always wanted to make a positioning device using an old style circular antenna as a radiogoniometer. Mounting the antenna on top of the servo or stepper motor would make a robotized location device to a beacon mounted in another device.
I don't have enough RF knowledge to make that, but having something like this would be pretty sweet.
By the way, back in time most of the sci-fi robots had a round antenna in the head 8)With a couple of antennas, if you just need low accuracy, you might solve it with RFID tags.
By circular antenna do you mean a magnetic loop? That's only going to be feasible at relatively long wavelengths. A directional antenna like a yagi would be more likely at ISM band wavelengths.
I think this could all get very expensive - perhaps ultrasound would do better?
Sovrac:
The devices will be moving in a random direction from each other, with physical interferences between them.When one has made a link with another (an arduino has enter the others range), they would have to know what direction the other one is in (to its left/right)
MOA could be overkill, would an RF transmitter coupled with another device provide a solution?
I think you need write more on the exact environmental situation, because right now people are guessing
or making assumptions about that will be.
For example what are the physical interferences? Is is a minor bump/short wall that they must go around or
are they full floor to ceiling walls that might be steel reinforced concrete.
And whats overhead?
Is it open sky or is this inside a building or other structure such that there could be no hope of ever using GPS.
Also there is an assumption that these are all remaining in the same vertical plane but you have
not stated that this is the case.
Things like this really matter.
So tell us the full parameters of the scope of the problem to solve.
--- bill
if you look at this video
they use some QR code type of thing and have an application running on a computer that looks for that Code/Picture and the robot navigates to it and waters the different plants accordingly to what code it has,
you could translate this into some sort of circular scanning camera looking for the arduinos in its vicinity , and you can mount some sort of QR code on each of the surrounding arduinos to be identified.. i dont know if this makes any sense but watch the video it might help understand what i am saying.....
its a tracking system , just that you would place the individual markers on the surrounding arduinos and the tracking application would be the middle arduino or the "seeker" ....
of course with Arduinos alone this is not possible, at least i dont know so do not take my word for it