Distance sensors for Trilateralization

[nwilkins259]

Was looking for a cool project concept involving a remote helicopter and I decided on trying to triangulate its location and maybe make it autonomous. Started doing the math and realized i was really looking to trilateralize its location using the magnitude of 3 distance sensors set in a triangle of length~10'.

What kind of sensors would be best to get an accurate distance measurement. I was thinking IR, or ultrasonic, or maybe calculating the decay in an rf signal. I was thinking a light receiver/transmitter on the helicopter as bouncing any kind of signal or wave off of the helo seems impractical.

specifications are
a large angle of efficiency (helicopter is a small mobile target)
range of about 10ft
small enough to not burden the helicopter

 

[Studiot]

Reflecting mirrors?

 

[skeptic2]

How would you determine the distance using IR? At 10 ft., (20 ft round trip) the time delay of arrival would be on the order of 20 nS. To get an accurate distance you would have to time the signal to better than 1 nS.

If you want to use decay (not delay) of an RF signal you will need to account for all the reflections in the area as well as the angle of the helicopter antenna to the transmitter.

Ultrasonic seems the most realistic to me but since the transducers are highly directional you may have to track your helicopter.

 

[Enthalpy]

Both RF amplitude and ultrasound are strongly impossible because of their erratic propagation. Ultrasound measures get possible in a protected environment (=a tube) and by filtering out all secondary signal paths.

The amplitude of a low-frequency (1-100kHz) magnetic field would work but this is seriously difficult.

Measuring within 1ns is possible but difficult. Converting a pulse width into a DC value is easier than a time counter. You could emit at the helicopter light modulated at 10MHz and compare the phases at 4 receivers. This is a long and difficult electronics project.

What looks easiest to me is particularly inelegant. Put a LED under the helicopter, several webcams with optical filters tuned to the LED, and add a piece of software to determine the direction of the LED seen by each webcam.

 

[schip666!]

I differ on a couple issues...

Ultrasonic distance sensors (try google!) can be used reasonably well up to 10 or so feet, depending on the environment. I don't know if using a bunch of them would cause too much interference, but many robot implementations do just that so they must be OK. Here's one sample: http://www.hobbyengineering.com/H2951.html

Also there is an IR distance sensor -- I don't know how it works -- that's good up to about 5', Sharp GP2Y0A02YK0F, c.f.: http://www.sparkfun.com/products/8958

have fun...