What Arduino board should I get to have a flying drone follow a signal?

Hi, I'm developing a small flying drone that would have the potential to follow an electromagnetic signal such as a radio wave being emmitted from a very small remote. It would have to connect to the main flight control unit to convert the data and command the motors/servos. I would also need infrared obstacle sensing boards but I have seem to have found those already. I hope to get a reply from someone soon. Thanks :)

-Mantas.M

The more important question is: how you will determine the position, or at least the orientation, of the "electromagnetic signal" with respect to the drone.

I agree with Jim. What you are proposing is not technically possible for hobbyists. There have been numerous other postings on this forum where people wanted to track people etc using some sort of beacon, and as far as I'm aware none of them have ever come to fruition. There are some fraudulent Kickstarter projects that appear to do beacon tracking on smartphones, but as far as I'm aware they have all been exposed as scams, as what they claim to be doing is not technically possible, normally because it breaks fundamental laws of physics.

The only way you could track a person or car etc, is for them to carry a GPS tracker that continuously sends its position. Via radio e.g 433mhz or 2.4g

mantas_m: Hi, I'm developing a small flying drone that would have the potential to follow an electromagnetic signal such as a radio wave being emmitted from a very small remote.

If the radio signal contains data to direct the drone (left-right-up-down, as appropriate) that may be possible.

If you just want to emit a radio wave of some (unspecified) frequency and detect where it is coming from that is altogether different and, I suspect, very difficult if not impossible for someone who has to ask "how?" here.

Think of the size of radar antennas on ships.

...R

Well, there is always ADF. en.wikipedia.org/wiki/Radio_direction_finder

Hmmm. I see where you guys are coming from. What I have in mind is that the drone is programmed to stay a certain distance away from the remote, as well as having infrared object sensors that stop it from colliding into you, and the objects around you.

How will it determine "distance?" Received signal strength is not a good indication of distance.

These types of posts come up from time to time. Measuring distance using radio is extremely hard to do for a hobbyist. You are basically making a a Radar system and such systems are beyond the capability of hobbyists due to the highly specialised radio gear thats needed. Measuring direction is aslo hard to do, given the physical size of the antennas that are used. Next time you visit a airport, look at the size of the radar antennas that are used to track planes.

Ive been pondering this for a while.

Im considering attempting to build a phased array antenna using ultrasonics (in air).

I know it works in solids.

Has this been attempted anywhwere ?

Basically an ultrasonic radar with an active transponder to get the range.

The most realistic way of achieving this would be to interface your arduino to a pigeon brain. It may take a bit of R&D to iron out the wrinkles.

I was thinking of a bat , but they are very difficult to catch :D

Boardburner2: I was thinking of a bat , but they are very difficult to catch :D

That'll be because you're thinking ultrasonics, whereas I was thinking magnetic fields. :)

The ultrasonic method should work but ultrasonic signals don't carry well. Their power is dissipated excedingly quickly over distance. An array of 3 antennae with a very short wavelength radio signal could possibly do phase comparison to get directionality, but it's no mean feet.

Have you started building the bat traps yet :)

Boardburner2: Ive been pondering this for a while.

Im considering attempting to build a phased array antenna using ultrasonics (in air).

I know it works in solids.

Has this been attempted anywhwere ?

I tested an underwater sonic direction finder, and it worked ok. The problem was echos can really throw off your reading.

I used three microphones (antenna?) less than 1/2 wavelength apart, and a "mixer" with a common local oscillator. The phase offset between mic inputs indicated the direction.

I experimented with this shortly using radio, and found the phase shift survived down to the Intermediate Frequency using a common local oscillator, if that helps.

Thanks iv experimented with similar using linear circuitry. Echos, yes , works outside but indoors or near a structure its hopeless.

Im wondering if several recievers and dsp might be the solution.

Its a point scource so processing should be nowhere near as complicated as the likes of sidescan sonar.

Did your system do ranging as well ?

Boardburner2:
Did your system do ranging as well ?

No, just direction. Strictly passive. You could use a transmitter in the receiver unit that would work like an aircraft transponder to incorporate ranging. The receiver would send a pulse that would cause the transmitter to send a pulse, and the time between the sent and received pulse would give a range. I haven’t tried that tho.

edit: A sonar mile in air is about 10 seconds. A sonar mile in water is about 2.25 seconds. A radar mile is 13.6 microseconds. This is the “round trip” of a transmitted signal, and does not include the propagation delay of a transponder.

KenTF: That'll be because you're thinking ultrasonics, whereas I was thinking magnetic fields. :)

The ultrasonic method should work but ultrasonic signals don't carry well. Their power is dissipated excedingly quickly over distance. An array of 3 antennae with a very short wavelength radio signal could possibly do phase comparison to get directionality, but it's no mean feet.

Yes difficult to get the information there.

figures I have seen comparing sonar. active sonar range 1 km passive sonar range 6 km for same equipment.

ultrasonic tape measure range 12 m so if it all compares well I would expect an active transponder system to have a range of 50 m or so.

did op say what range he wanted

main problem I see is getting the transponders.

smallest I have are about 2 cm dia and the maths says they should be spaced under 1 wavelength apart

KenTF: Have you started building the bat traps yet :)

needs work, keep getting this annoyed bloke in black underpants

:fearful:

Scary home work here

http://www.personal.psu.edu/faculty/m/x/mxm14/sonar/beamforming.pdf

Chip that does it all here

https://www.microchipdirect.com/ProductDetails.aspx?Catalog=BuyMicrochip&Category=MD2131&mid=12&treeid=3&lmid=2082

Boardburner2: Scary home work here

http://www.personal.psu.edu/faculty/m/x/mxm14/sonar/beamforming.pdf

I think that's probably over complicating the issue. Rather than attempting to gleen exactly which direction the transmitter is. All that is needed is some feedback of whether it's to the left or the right.

So if we had just two antena, we could then use the output of the phase locked loop to determine whether we're on the correct heading. If the output of the PLL shows unity, we keep going in a straight line. If however there's a phase difference, we just turn to the left or right accordingly.

The problem with this is that our drone would be just as happy flying directly away from the target as towards it. Hence the need for a third antena.

BTW. If you approach your bats during the day, you won't even need to creep about. They are totally oblivious to the noises within our normal hearing range and will be fast asleep.

KenTF: So if we had just two antena, we could then use the output of the phase locked loop to determine whether we're on the correct heading. If the output of the PLL shows unity, we keep going in a straight line. If however there's a phase difference, we just turn to the left or right accordingly.

The problem with this is that our drone would be just as happy flying directly away from the target as towards it. Hence the need for a third antena.

That is how I started. Two antennas. It kinda works if you do not need to know the heading to the transmitter. It does require the two antennas within 1/2 wavelength tho. Like you said, if the PLL/mixer is locked in phase, the drone will not know it is 180 degrees out without a third antenna or making a turn. And even then, I found the maths easier if the antennas are still within 1/2 wavelength of each other.

My detectors are big so the closest i can get them is several wavelengths.

This causes sidelobes and trying to determine the correct lobe by amplitude is very difficult.

Bats, cant find where they sleep , bloke in underpants has faster car than me.