Approximate distance from RF

[airman00]

I guess my idea of sending multiple signals at different power levels was correct.
I emailed Wild Planet , those makers of the Spy Tracker Toy - http://www.amazon.com/Wild-Planet-Gear%C2%AE-Agent-Tracker/dp/B0007V62XQ

They replied with the following:

Dear Eric,
I can offer you this, from our engineer:
The transmitter sends out 3 different RF signals; Close, Medium and Far. Each signal has a maximum distance that it can be detected by the receiver unit. The receiver picks up each of these signals and filters out all but the signal that corresponds to the closest distance. It then indicates this distance by activating a light and sound.
I hope that helps. -Kim

I'll be trying out this method of RF approximation very soon

[airman00]

Heres a schematic of the power selector for the transmitter power
There are transistors which are all connected to pins on the microcontroller. The microcontroller can control each switch individually .

There are diodes on top with a voltage drop of 2V. Therefore after each diode the voltage is lowered by 2V. Microcontroller switches the transistors to select voltage at transmitter.

Using this pseudo code on the transmitter :
Set to 12V
Send Byte (12)
Set to 10V
Send Byte (10)
and on and on until 2V

Then the receiver just checks to see which bytes it gets and can therefore determine range.

Any feedback or comments?

[Admin]

I think your idea will work . . . but you'll probably have an accuracy of about +/- 15 feet. Not very useful for an inside robot.

For an outdoors flying robot, I might actually try this myself!

[airman00]

I think your idea will work . . . but you'll probably have an accuracy of about +/- 15 feet. Not very useful for an inside robot.

For an outdoors flying robot, I might actually try this myself!

I did some tests and it worked fine. I did outdoor tests but haven't really checked the accuracy of it., just checked to see if it more or less worked.

In the end I used RSSI-equipped RF modules instead of this method.

[Asellith]

just a thought I had while reading this thread.....

Your using omnidirectional antennas right? Why not use a directional antenna. might be expensive to buy or build but if it is designed right you can setup several transmitter locations and then scan with a servo. Reflection will be cut down because your looking for high points only. Find the three max signals and then use those bearings to calculate the relative location. now this might not work in your situation because the transmitters need to have defined locations in the memory of the robot. Well some of them need to be. If you can detect and decode an identifier for each transmitter then you can have at least 3 located in specific locations and one mobile. Then the robot should be able to calculate the location of itself in the space and then the distance to the mobile transmitter.

I remember trying to figure out a way to locate a swarm of robots with 3 robots having GPS and the rest calculating coordinates from those three defined locations.