IR location with beacon

I am looking to have an IR beacon place on a person sort of like a necklace. I have a robot that I would like to face the person based on which direction the IR beacon is located. I was thinking I would need 12 IR sensors all placed in a circle on top of the robot and have the robot sense the signal strength of each sensor and turn the robot to that direction.

what kind of sensor do I need to do this?

Thank you for your help.

What happens when the person is facing away from the robot? A standard IR detector module should suffice for the detector.

I don't mind if the person turns around. What do you mean by standard IR detector module? can you give me a link to a product?

What do you mean by standard IR detector module?

The usual 38kHz remote control receivers, I guess.

how could I hookup 10 of those to the Arduino?

One per pin?
Trouble is, IR reflects, so direction would be tricky.

Hi, To make this directional you need to mount a receiver in a black tube etc. to restrict the direction.

Then you could use a servo to point it and read the signal strength.

I put an ultrasonic ranger on a pan-tilt like this: Sensor pan/tilt kit with 2 Servos One of my grandsons built a robot with this that "looks around" as it is moving. Here's a photo of an earlier version that used a single servo to point the sensor around:

DISCLAIMER: Mentioned stuff from my own shop...

IR LED’s respond quite well to variance in light levels. And they are cheap so you could mount quite a few, although reading LED sensor data with an arduino requires either 2 digital pins per LED or 1 analog pin per LED.

There are two ways to use the LED as a sensor; capacitively, or with an amplifier. —>google<—.

As Terry suggested you would need to mount each LED inside a tube to control were it receives its light from.

I would think a black tube would be a bad idea since IR is basically heat and black would absorb heat from the environment (light), and skew the sensor data as the tube heats up. Use a white tube.

I would think a black tube would be a bad idea since IR is basically heat and black would absorb heat from the environment (light), and skew the sensor data as the tube heats up. Use a white tube.

No at the sorts of IR wavelength we are talking about here it is not affected by environment heat.

A standard IR receiver for remotes will not work because you can't get signal strength from them.

No at the sorts of IR wavelength we are talking about here it is not affected by environment heat.

If he were to use IR LED's as I suggested they would respond to heat, and also to Red light.

If he were to use IR LED's as I suggested they would respond to heat, and also to Red light.

No they would not respond to heat.

IR LEDs work at 850 to 950nm this is Near-infrared.
Heat is detected in the Long-wavelength infrared wavelengths.
see:-

should I just get an analog IR LED receiver or should I use some sort of pulsing?

Yea, using LED's as sensors by measuring capacitance requires some switching and measuring of discharge times which could become troublesome if one or more of the LED's is receiving no light. Better to amplify the voltage change created across the PN junction when light hits it.

IR LEDs work at 850 to 950nm this is Near-infrared.

The LED emits light in this range, but other wavelengths will affect the PN junction in the diode to a lesser degree. so it may respond to radiation in the micrometer range.

IR sensors (that look like LED's) usually have a dark plastic casing that filters out everything except the IR in the range they want to measure.
Whereas the IR LED's usually have clear casing.

should I just get an analog IR LED receiver or should I use some sort of pulsing?

Having a receiver that looks for a pulsed signal (like regular IR remote controls send) is important to get a good signal-to-noise level. It also discriminates against constants like sunlight, room lights etc.

Take a look at the http://ArduinoInfo.Info WIKI on this: look at SENSORS > IR REMOTE CONTROL Also follow that to Sam Bergman's great page on IR for details.

You need to have the "Beacon" keep sending signals that are seen as valid IR transmissions by the receiver. You can set up several IR LEDs as the transmitter (Infrared LED 5mm (Package of 10))

You can just set up some receivers with 5V supply and an LED/resistor from +5V to output and walk around with any IR remote and point it at the receiver. You should see the LED flash when a signal is received.
Receivers are cheap (http://yourduino.com/sunshop2/index.php?l=product_detail&p=210 and you can set up several to test.

Here's the directional charcteristics of the typical IR receiver. As you can see, it is intended to work when the transmitter is off-axis up to 45 degrees or more (like on your TV)..

Here's an example of a pulsed transmitter and receiver pair: http://arduino-info.wikispaces.com/InfraredBeamPair which works well over some distance even in relatively bright light.

Pololu has a IR Beacon transceiver system for about $50 which is a good example: Pololu IR Beacon Transceiver Pair and you can get some good ideas (Or the Beacon system) from them. You can see Christopher Hazlett's Robot design HERE: robotishappy.com

If you get something working, please post back here with examples of what you've done. I'd be happy to publish that in the Projects section of ArduinoInfo.

DISCLAIMER: Mentioned stuff from my own shop... but you can get all this stuff from Adafruit, Sparkfun or Ebay.

but other wavelengths will affect the PN junction in the diode to a lesser degree. so it may respond to radiation in the micrometer range.

No No No.
Your grasp on semiconductor solid state physics is very ropey. A photon of radiation has to have energy greater than the gap between the conduction band and the valance band to excite any electron into a sufficiently high enough energy state to take part in conduction. The energy in a photon is dependent only on its wavelength. Any silicon based detector will NOT detect any radiation associated with heat at temperatures outside that of a furnace.

should I just get an analog IR LED receiver or should I use some sort of pulsing?

You will not be able to get very much range with a normal IR transistor. The types for remote controls need a pulsing source so that the amplifier is not saturated by the ambient IR light. However, they have a limit to how long they can receive pulses without giving up. This is something like 20 pulses maximum. There is only one IR receiver I know of that is not built like that and it is the TSOP4038. As the name implies it needs a 38KHz pulsing IR source in order for it to detect. There is no signal strength information available just a straight forward detect / no detect.

Like many have said that the "normal" IR module would not give you signal strength so you can't locate the source based on that. You can mount an IR sensor on a motor, maybe with some kind of rotary encoding, rotate the IR module like a radar... and if the IR source on necklace can be made to pulse at 38Khz (or the frequency of receiver), then you can use the "normal" IR module so you can improve upon N/S ratio, as well as taking advantage of all circuit (filter, amplifiers, etc) built into the IR module.

just a thought.

Sunlight on a wall can be picked up by simple near IR detector. The pulsed kind can handle a certain amount.
If it shows white on a CCD cam (web cam, phone cam, etc) and doesn't look white to your eyes then it's near-IR overloading the detector. A red led puts out strong near-IR light.

I suggest you use a pair of IR photodiodes. If you mount them facing in different directions by (say) 90 degrees, and you choose a photodiode with a half-angle view greater than half of that, then the robot is facing you when the signals from both photodiodes are equal. If you use 4 photodiodes facing at 90 degree intervals, then the robot will also be able to tell which way to turn. With just 2 photodiodes, it may have to try turning both ways to see which way makes the signals more even, and reflections may be a greater problem.

As has already been said, you will need to modulate the IR to distinguish the transmitted signal from background IR. You will also need to amplify the signal from each photodiode and filter out just the signal you want. See “Multiple feedback band pass filter” at http://www.electronics-tutorials.ws/filter/filter_7.html for one possibility. In practice, the sensitivity of the photodiodes may not be quite equal, so you may have to trim the amplifier gains to balance them.