Society of Robots - Robot Forum
General Misc => Misc => Topic started by: nathan25 on November 23, 2007, 11:48:41 PM
-
Which would you put on your robot. the IR sensors or the Photosensors. I'm having problems with the photosenors and i want to know if it would be worth it to switch. Tell me why i should.
Thnxs
Nate
-
A little something I read from http://www.robotroom.com/ColorSensor.html
Speed Limitation
Only two or three complete color readings can be performed per second. The cadmium-sulfide photocell needs time to react to the brightness changes.
So switch to IR if you need to be fast.
-
Depends a lot on what you want to sense, environment, algorithm, etc :P
-
Light sensors can't give environment data that you can rely on. If it's light, it doesn't have to mean there is nothing in the way (a lamp or a very white surface for example). IR however always give the same data, whether or not there are light or shadow.
If you want a robot that can avoid obstacles you should go for an IR sensor. You can still have light sensors however, and make it attract to lighter areas and still being able to avoid obstacles correctly. I recommend you buying an Sharp IR sensor rather than building a sensor yourself, because that way you can get accurate distance data.
-
IR however always give the same data, whether or not there are light or shadow.
Thats not necessarily true , sunlight does affect IR readings.
It all depends what you want the sensors for ; obstacle avoidance or line following or color
I would suggest IR for all three, except obstacle avoidance where I am partial to sonar.
But i've used IR for ranging and it worked fine ( but sonar was better :D )
EDIT: Also remember to enclose the IR sensors , as not to screw up the readings.
-
IR however always give the same data, whether or not there are light or shadow.
Thats not necessarily true , sunlight does affect IR readings.
Yep, i know. Just thought it would confuse, because they work in totally different ways. IR uses its own ray as reference, and doesn't sense the amount of light but the angle (when speaking about the Sharp IR sensors that is) the light is reflected at.
Here you can find a lot of useful information about IR sensors: http://www.solarbotics.net/library/circuits/sensors_prox.html (http://www.solarbotics.net/library/circuits/sensors_prox.html)
-
Despite what people believe, I discovered that sunlight actually does affect readings of the Sharp IR. They are resistant to environmental light, but not immune.
An urban legend in robotics dies today :P
-
I know about sunlight affecting it - even if it is indirect
But would fluorescent or incandescent lighting have affects on the readings?
-
But would fluorescent or incandescent lighting have affects on the readings?
If your lightbulb gets warm/hot, that means its emitting infrared light, and as such interfering with your sensor.
But I'd say the interference would be minor - just run an experiment and see how it affects your reading.
-
Yes ,but then the experiment would have to be in an extremely dark room ( besides the backlight of the LCD to see the readings)
What I've found is that if the room is filled with sunlight ( meaning it has windows and its a sunny day) , the IR reading would be slightly affected. Thats why you would have to cover it to prevent any light from coming through.
If your lightbulb gets warm/hot, that means its emitting infrared light, and as such interfering with your sensor.
LOL , i forgot that heat emits IR light . Its been a while since I played with heat / human detection.....
-
my understanding is that the sharp's IR is modulated to help resist enviromental noise such as the sun/hot lights or lamps. In my experience with them they are fairly accurate outside in phoenix sun during the summer BUT when reviewing some of the data's log I see occassional spikes and such. I never really saw situations where it would consantly off and make an "artifact" in the readings..
With some mild programming knowledge/common sense you could easily filter out most of the noise.