IR Sensors- will they attack hot Chiaphuas?

[Gabriel]

Thank you for pointing this out! I don't know the answer, but
at some point I'm definitely going to ask the filmmaking people at my college to bring in some cameras so that we can test this.

This is my first year as a team leader and I feel a little like the guys on Apollo 13. There's a line from the movie where one of the astronauts (Swaggert I think) is concerned that they will overshoot the Earth. Lovell replies "there are about 1000 things that have to happen for us to get home. We're working on number six, you're worrying about number 520." Its just one more item on the giant list of things to worry about

[KyleGilbert45]

One other thing I've thought about. What about the Night Shot on some camcorders such as Sony. Are those IR devices? Even though you wouldn't be using it at a competition, could someone use it to knock a robot off it's course?

[Mike AA]

To put it simply THE INFRARED SENSORS WILL NOT DETECT ANYTHING OTHER THAN WHAT THEY HAVE BEEN DISIGNED FOR! ie. AN INFRARED EMITTER EMMITING AT THE FREQUENCY THE INFRARED DETECTOR HAS BEEN SET AT. I worked for 3/4 years designing and programing a pinewood derby racetrack finishline that uses infrared. We have not had any problem with extra light or other types fo thigns that interfere.

[pauluffel]

As these sensors on the robot will be much more receptive (as they're designed to seach a wide area for the beacon) I think they would be much more prone to interference (I remember at the FLL Competition there was trouble with all of the people downloading programs with IR to the Mindstorms at the same time and accidentally downloading over other teams' programs and the overhead lights interfering somehow) but since these detectors are designed to pick out a specific pattern, I think the interference will not be a problem unless someone in the crowd has a beacon (since we're to be given the specs for a beacon so we can make one for testing).

Also, since picking out the proper pattern of a beacon is a programming thing, I see the potential for someone to make a custom beacon and program their autonomous to follow that around so someone could remotely control their robot during the first 15 seconds, but I think we're just going to have to trust in GP that no one will do that.

[Greg]

Originally Posted by Mike AA

To put it simply THE INFRARED SENSORS WILL NOT DETECT ANYTHING OTHER THAN WHAT THEY HAVE BEEN DISIGNED FOR!

It depends on the quality of the sensor really. I have experimented with building a LIRC-style universal IR receiver once and I can say that some detectors are actually affected by direct sunlight and fluorescent lights. Others are not affected as much. However, even if this does happen, the demodulated data is garbage, and will never be equal to the signal sent out by the beacon. So the robot will more than likely never consider a camera or a hot motor a beacon. However, strong interference may not allow a bot to recognize a beacon. This should not be a problem, however, since the playing field is usually quite dark.

[Anthony Kesich]

if it is a problem, you can get some aftermarket IR sensors that give back an analog signal, not just YES/NO. Then you can figure out what is strongest and most likely the beacon. And if you have a good programmer, with sensors attached to servos, he can make the bot check to make sure the IR it is detecting is comming from the right direction relitave to your starting position.

[Greg]

I dont think this is possible. As has been said before, the signal is modulated. It is not just a "light on/light off" thing. The beacon is likely sending out modulated binary data, and the receiver gets the pulses, demodulates them, and then the software compares the binary signature. This is how a TV remote works. An analog detector would just get the average brightness of a pulsating source. Of course, all of this is unconfirmed - I have not seen the code yet and all the IR stuff is missing from our kit

[Anthony Kesich]

ahh, good point. I'm used to constant signals, but who knows? I think some experimenting is in order. Maybe some interrupts to collect the strength for 2 ms... hrmmmm

-Kesich

[Curtis Williams]

Wouldn't these things be as realiable as the sensors from last year we used to detect the reflective tape? I think they worked the same way.

[Anthony Kesich]

Didnt those use the visible light spectrum? And the light sensors projected, whereas the IR sensors just pickup, so there really isnt the chance of seeing reflection when there isn't one. I went straight dead reckoning, though. Getting to the top of the ramp and hitting boxes in 5 flat seemed better to me. Anyways, in the line vs IR arguement, i'd say IR because you can see the IR from anywhere. The line, on the other hand, can get smugged up, lost, and once you are done with it, the sensors are useless. Ir can triangulate though.... hehehe

-Kesich

[Adam Y.]

The signal from the beacons and receivers probably resemble the signals from a television remote. I have never encountered any interfernce from body heat on those things.

[Anthony Kesich]

Wouldn't that be funny. Walk by the TV with a candle and the tv turns on, mutes, unmutes, cranks the volume, then hacks the cable box and gets you free PPV for life. Heheheh :-D

-Kesich

[Al Skierkiewicz]

OK,
Before you guys get too crazy, the IR emitters generate energy in the same area as heat but have one distinct difference. The emission will occur as 40kHz pulses for varying lengths of time. That is, bursts of 40kHz modulated IR energy. The night illuminators on cameras are just that, steady state, non modulated sources the same as hot bodies. The IR camera range finders and remote controls use modulated IR emitters at different frequencies so there should be no real problems if you keep in mind that you are looking for a 40kHz modulated source. The bandwidth of the IR is not 40kHz but likely is much wider than that since the bandwidth you are measuring is near visible light.

[Kevin Watson]

Originally Posted by Al Skierkiewicz

OK, Before you guys get too crazy...

Too late .

Originally Posted by Al Skierkiewicz

That is, bursts of 40kHz modulated IR energy. The night illuminators on cameras are just that, steady state, non modulated sources the same as hot bodies.

Yes, as Al correctly (again) points out, the Vishay TSOP4840 sensors that are included in the kit and also used for the kick-off demo specifically discriminate against these type of energy sources. They also discriminate against non-modulated 40KHz sources. I've put a link to the data sheet here: http://kevin.org/frc. The modulation scheme is discussed in this thread.

-Kevin

[Joe Beynon]

this is something that i asked about at the main kickoff because of the fll problem. the guy told me that he tested the ir sensors with 5 cameras that were said to have problems with the fll bots and none of them caused interference. he also mentioned that these ir sensors are the type used in high-end stereo equipment and the only thing to possibly worry about would be a flickering flourescent bulb.