Wireless Beacons

For a project I am attempting to make wireless beacons to determine the location of a robot through trilateration. I would be using infrared to find the distances between the beacons and robot, with one infrared light on the bot and the beacons all acting as receivers. I had planned to use RF trancievers to communicate between the two, however I'm not sure how to interpret and send the analog reading from the beacons to the Arduino. Trying to find a cheap solution to this, any help would be appreciated.

More detail required...

  1. Is the robot's position determined by IR on three relay stations, which then transmit data by RF to another Arduino?

  2. A beacon transmits something but yours receive, would my use of "relay station" be a better description?

  3. Is this for indoor/outdoor use?

  4. What range do you need?

  5. Do you mean localisation rather than trilateration? Trilateration implies that the absolute distance can be determined, IR isn't any good for that, nor anything else within a reasonable budget for that matter.

  6. Is your question about which RF module to use, or how to send data by RF?

HINT: You question is unclear, read the Forum Posting Guidelines and spend some time adding more detail. There are quite a few RF and IR experts on the forum, a good question will get a good answer.

With neither IR or RF is is possible to determine distance over such a short scale (notably, the scale within which you could implement IR communication) as the trip time, at the speed of light, is far too small for simple equipment to measure.

So how would you propose to do this? Directional antennae scanning back and forth? Sweeping a narrow IR beam would perhaps be more practical?

Paul__B: With neither IR or RF is is possible to determine distance over such a short scale (notably, the scale within which you could implement IR communication) as the trip time, at the speed of light, is far too small for simple equipment to measure.

So how would you propose to do this? Directional antennae scanning back and forth? Sweeping a narrow IR beam would perhaps be more practical?

Having a narrow-beam spinning light on top of the robot and getting pings from the beacons via RF when they see it, you could determine position from angles, not travel-times.

NRF24L01 is an option. Haven't tried it myself, but have some lying around. Those are simple serial transceivers, and other people on this group say they get stable communication with these. They are quite cheap.

How do you propose to determine angles? how will you account for transmission delays and interrogation delay because of single threading in your code?

Interesting problems. Good luck.

123Splat: How do you propose to determine angles? how will you account for transmission delays and interrogation delay because of single threading in your code?

Interesting problems. Good luck.

Stepper motor on top of robot, rotating light source or angled mirror.

Shared clock between robot and beacons.

RF comm initiated by robot, querying beacons as to when they last "saw" the robot.

RF comm also used to maintain synchronized clocks.

Now, getting accurate readings on those beacons, is a different matter. The "light source" may have to be modulated in a certain way to be recognizable at the beacons.

It should be doable, and this is one of several options that I considered earlier, but I rejected it still. Too complicated.

Stepper motor on top of robot, driving a light source, or an angled mirror.

Shared clock between robot and beacons.

RF comm initiated by robot, querying beacons as to when they last detected the robot light source. Correlate with rotational history of the stepper, voila you have angles.

RF comm also used to maintain synchronized clocks.

Now, getting accurate readings on those beacons, is a different matter. The "light source" may have to be modulated in a certain way to be recognizable at the beacons.

It should be doable, and this is one of several options that I considered earlier, but I rejected it still. Too complicated.

Complicated is putting it mildly. You still have not addressed transmission delays (NRF2400's end up doing a LOT of retrains because of missed ack's).

Just a thought: If the bot knows the placement pattern of the beacons; The beacons are true beacons and re-transmit modulated broad beam ir message consisting of "start, beacon ID, end, pause..."; the bot has a 360 degree servo synced to 0, or 90, or whatever degrees is forward direction of bot, mounted on top of bot with a narrow band ir recvr slaved; bot senses a beacon signal ; bot knows what relative bearing the received signal is coming from; bot saves bearing of captured beacon signal; bot hunts for signals from adjacent beacons; bot captures adjacent beacon signals at determined angles; with a little algebra, bot calculates approximate location in field covered by beacons.

forget the RF.