Visible light sensors for line follower

Question

I'm making a second version of my line follower robot and I'm thinking what are the line sensors going to look like.

In the previous version I've successfully used IR LEDs and IR phototransistors, but I've heard stories about lines printed on an inkjet printer not being detected by some IR sensors and that got me started thinking about basing my new line sensor on visible light.

• I think that the sun is going to be pretty strong in the near IR too and I'm subtracting ambient light from the signals anyway.
• Rangefinders on cameras and other robots all use some IR wavelengths, so I'd be affected by these too. In visible light I'll at least have a chance to see the noise.
• If I see a line as black with my eyes, this means that a visible light based sensor will be able to follow it. In IR this is not so clear.
• Easy to determine if the sensor works or not.

What are the downsides of using visible light?

Answer 1

The Danish line follower robot competition "DTU RoboCup" recommends using visible light sensors in their competition, since IR sensors have a hard time distinguishing the floor from the line. I've participated in this comepetition, and seen quite some IR-using robots getting confused and wandering off.

My experience is that using visible light has one major downside that has to be dealt with. You mention it yourself: Sun light/daylight/ambient light. And since you're already dealing with that, I think you're covering the single most significant downside already.