Robot Communications —How to talk to your robot. (Page 1)

The picture above is a 2 Watt Blue Laser I built several months ago. It's way too powerful for robot communications around the house (it would start fires), but it looked like a good enough illustration for this page.  :-)


All right, this is intended to be a somewhat comprehensive write-up concerning different ways to talk to your robot. Whether it will be all that comprehensive remains to be seen, but let's get started.

Okay, first up, let's investigate infrared. We humans cannot see infrared, but that doesn't prevent our robot from doing so. Some of the microcontroller chips already come equipped with the ability to decode infrared signals from something like a TV remote. (I'm thinking of picaxe here.)

However, what I want to show is simple schematics for circuits you can easily add to your robot. Let's call it modular design. First, here is a very simple sketch that shows how to drive an infrared LED. The input marked with the letter S must be driven with a different circuit such as a microcontroller. The benefit of using this circuit, is that it pulls much less current from the microcontroller output pin (Specifically, only half a milliamp).


But what if we want to make one or more of these infrared beacons, without using additional microcontrollers? Isn't there a cheaper way? Yes. I have drawn a couple different possibilities, which you could use on your robot. Here is a circuit using a 555 timer chip.


"How can I use that to send data to my robot?" That is an easy fix. One way would be taking the 555 reset lead off of V+ and switching it on and off with something capable of sending data, such as a microcontroller. (All you need is a small one like an 8 pin chip). The data or code coming out of a microcontroller would be input at the lead tied to pin 4 of the 555, with the 1 KΩ resistor there to limit the current. [Another alternative for inputting a modulation signal would be to leave pin four tied to the positive supply voltage and using pin five instead.]  This was not done because removing the capacitor from pin 5 might affect the stability of the oscillator.

 The 555 chip is nice inasmuch as it will drive several LEDs. What if you don't have a 555? Do you have a logic gate like a 7400, 7402, 7404 or others? There are a lot of different types such as 74LSxx, or 74HCxx and others. Don't let that confuse you. Any of them will work in the following circuit.


And to wrap up this page showing IR or Visible light signalling. here is one more pulser circuit and a detector circuit using the 38 kHz detectors which are available many places for low cost.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
lumi's picture

You always come out with such information when I need them ;-)

Thanks for that value info.

BTW, i built one of that 555 beacons and realized that the frequency changes with the supply voltage. How to make sure that this is not happening?

Dan M's picture

This sounded odd. The 555 was designed to be immune to differences in supply voltage. To double check this I went to my library and dusted off my copy of "The Linear and Interface Circuits Data Book for Design Engineers" by Texas Instruments. Here are a couple excerpts from the 555 section:

    "The frequency of oscillation may be determined by the external resistors Ra and Rb and the external capacitor C. Duty cycle, D, is determined by the values selected for Ra and Rb and may be calculated as:  D = Rb /  (Ra + Rb) "

    "The duty cycle may be controlled, therefore, by the values of Ra and Rb. The capacitor C will charge through Ra and Rb and then discharge through Rb only, connected between the threshold (pin 6) and discharge (pin 7) pins. The astable connection results in capacitor C charging and discharging between the threshold voltage and the trigger voltage. The charge rate of the capacitor C is then controlled through feedback at these two pins. As in the monostable circuit, charge and discharge times (and therefore the frequency and duty cycle) are independent of the supply voltage."


Then, even though it said it was independent of the supply voltage, at another spot it says:


     "The operation of the timer may be modified by modulating the internal threshold and trigger voltages. Decoupling the control voltage input (pin 5) to ground with a capacitor may improve operation."


They show placing a .01 µF capacitor from pin 5 to ground.



lumi's picture

Thanks Dan, yes this is what i know too. It would not make sense when the circuit would be affected by the supply voltage. 

I was experiencing this behavior by a change of the supply voltage from 9V to 5V. However, maybe i should check tonight if there is something wrong with my circuit.

birdmun's picture

The 38khz receivers should be unaffected by minor voltage drops. If you were to come up with a way that you knew the sensor should pick up the light, you could use it as a test case. If you had a spare INPUT, you could use an interrupt and a timer/counter to measure the frequency and know if it had dropped out of the useful frequency range. Or, you could measure the battery voltage, and then, you would not only know when your frequency was off, but, when you needed a recharge.

lumi's picture

Thanks bird, but the beacon should be a stand alone device without any µC. Because of this we used just the 555. 

You are right, minor drop of voltage is ok, but i was thinking about from 9V down to 5V. Without adding a voltege regulator (costs) I just have to be aware of that.

Maxhirez's picture

I want to see you arm Schroedinger with that blue laser.

Where do you even find a 2W 450-500 nm diode? Wow...

Dan M's picture

I actually considered mounting it on Schrödinger, but decided it was too dangerous to give an autonomous robot a laser that dangerous. I figured it could be ME that got my eyes blinded by it.

birdmun's picture

pulsed blu-ray burner?

Dan M's picture

Well, let's see... Here is one for $856.94 with shipping.


ha ha

but you can bet I didn't pay that much.

I got one off ebay.  I forget how much, but I think I might have paid as much as $40 for the diode with shipping.

jkeneally's picture

Well written and documented!  Excellent resource for expert and novice alike.  I look forward to future installments.