Go Down

Topic: IR sensor and emitter advice please. (Read 633 times) previous topic - next topic

LennyLemming

Hi there. I am completely new to the world of Arduino and have a project I would like to get underway. What I need is some advice on suitable infra red sender and recievers so I can begin experimenting.

Here is my problem. I want to hang a computer monitor from a counterbalanced arm that is pivoted on a pan and tilt mechanism driven by motors and positioned above my head. The idea is that if I had four infra red sensors attached to the monitor top, two for up down, two for left right in a diamond shape and had an infra red source on my head then when I moved my head the monitor would move and keep in front of my face.

The aim being if one sensor of a pair recieved more light than the other the monitor would centre itself so that both would receive equal light. The distance from source to sensor would be between around 300mm to 500mm.

Any suggestions, please? Also book recommendations.

I have ordered a kit with Arduino board, breadboard and motor shield plus leads and USB cable and have a couple of suitable 24 volt motors and a power supply.

In case you are wondering I am a bit of a combat flight sim addict. Normally you would look at a stationary screen and the view would slide around behind it which is a little un-natural. Head mounted visors are prohibitively expensive and have other disadvantages. I would also use rotary encoders (or hack an old ball type mouse) and attach them to the xy axis of the pan tilt mechanism so that the view would move in the opposite direction to the monitor. Thus the view should appear to remain still and it would feel like I am actually looking around.

Nikarus

Not gonan lie itd be a lot easier to just get a couple other monitors to do this. But hey I'm a mech sim guy adn I've got my command chair too so more power to ya.

Now pretty easy and cheap IR recievers and emitters are https://www.sparkfun.com/products/241

Now your first goal with this project would probably be jsut getting your IRs to actually sense and send a feedback signal.
This However isn't waht I'd personalyl recommend. I'm goign to assume you've seen the Ted Talks WIImote hacking. The thing is is that you dont need jsut a couple IR sensors. you need tons of them to make up an IR "camera" to find your dot on.

The Wiimote head (theres guides about it around) has an IR camera/processor thats able to track up to 4 IR sources at a time (put the IR Emitters on the top bottom and sides of the screen instead of recievers). Then use the data streamed out by the Wiimote to track where the monitor is relative to your head and the direction your facing.

This instructable might give you a bit of insight into what your trying. The guy makes a servo move a the camera to look at an IR LED

http://www.instructables.com/id/Wii-Remote-IR-Camera-Hack/step6/Using-the-camera-to-control-something/

LennyLemming

Thanks for your reply, Nikarus. I have looked at the Wii camera hack and it is a possibility along with Freetrack or TrackIR. Initially I was looking for something really simple and cheap. I have checked out such things as line following robots, robots that home on IR beacons and robotic tanks that aim at an IR target. I just need to get some bits and start playing.

I am also thinking of light dependent resistors. You are supposed to be able to get ones that operate in the IR spectrum but I have not found one yet. Visible light ones might work though with a pencil beam and a shroud to keep out most of ambiant light.

I am busy researching and was hoping for some pointers given the application and distance involved. The Society of Robots site I found checking your link to Sparkfun is looking promising. Thanks again.

Ps. A multi monitor setup still does not solve the issue of looking at a fixed point whilst the view moves behind but such a set up on a pan and tilt mechanism zooming around your head would be awesome.

Nikarus


Thanks for your reply, Nikarus. I have looked at the Wii camera hack and it is a possibility along with Freetrack or TrackIR. Initially I was looking for something really simple and cheap. I have checked out such things as line following robots, robots that home on IR beacons and robotic tanks that aim at an IR target. I just need to get some bits and start playing.

Well the IR sensors and emitters I linked you are easy enough to hook up and start playing with. The thing I can see as a problem with your idea is it will be difficult to track what angle your head is relative to the screen. But I haven't given that kidn of system that exceptionally much thought (since the motors probably wouldn't be able to react to your head motions quite as fast as you'd want them to.

I am also thinking of light dependent resistors. You are supposed to be able to get ones that operate in the IR spectrum but I have not found one yet. Visible light ones might work though with a pencil beam and a shroud to keep out most of ambiant light.

Photoresistors could work to. These are best used for sensing the intensity of light though. I could see a lot of issues coming up with tryign to use them to track a particular light however.

I am busy researching and was hoping for some pointers given the application and distance involved. The Society of Robots site I found checking your link to Sparkfun is looking promising. Thanks again.

Given the distance your working, most any of the IR setups will work. Actually for airsoft I've tried putting IR strobes into the shoulders of a couple of my teammates and making it so that when I point my gun at them, the gun detects the strobe and will blink the red dot sight to show that its one of my friends I'm aiming at (within about 3 degrees accuracy too with the tube I setup to block out other angles). And when using an array of IR leds on each shoulder, I could get them to sense up to about 70ft away (far enough that through the trees I couldn't tell they were on my team) However that was a couple dozen LEDs For head to monitor, of about 3 ft range, you're definable gonna be fine for getting a good signal. Make sure to watch out for viewing angles on your detectors though.

Ps. A multi monitor setup still does not solve the issue of looking at a fixed point whilst the view moves behind but such a set up on a pan and tilt mechanism zooming around your head would be awesome.

Back to the zooming. Thinking fast It wouldnt be to expensive to hang your monitor from a rotating jig from the ceiling (with a counterweight). And with good bearings and a sufficiently powerful motor, It could probably respond fairly quick, Hope noone is standing nearby at the time. Now this mount would rotate the monitor around your head. and probably max out at going 90 degrees either way (since it would be hard to make the slip rings to do a 360 rotation). Pan and tilt motors wouldn't be too hard to manage cause they wouldnt need to be anywhere near the size. The lift and lower motor though I can see trouble.


Though I might just be blowing it out of proportion and you're just planning on panning and tilting the monitor so its always facing you, without it moving its base at all. Well thats easy then, the other ones hard but still doable.

If you could actually give a little more insight into what you're actually trying to build (quick drawing perhaps) I might be able to be a bit more helpful, cause I'm pretty sure what I'm envisioning is wrong. Best of luck to you though either way.

PS. which sims you play if any of them are online? Its a lot of trouble to get good matches going in a lot of games. And I'm always looking for a good fight.



LennyLemming


No, I haven't given up.

Doing a quick drawing showed having the pan and tilt above my head wouldn't do. It would need to be as close as possible to the neck where the head turns and tilts. This is still perfectly workable though and have attached a couple of rough sketches to give idea of what I am aiming at.

I pretty much exclusively play the IL2 Sturmovic series of sims including Cliffs of Dover. Great fun on line but enormous learning curve.

I have been collecting together and ordering the bits to make the mechanism which I shouldn't find a problem and then will have something more concrete to test the IR sensor side of things on. If it dose go a bit quiet I will send a message when I am further forward.



[/img]



Go Up