Re: Barcodes & error detection

In lugnet.robotics, Philippe Hurbain wrote:

> I had in mind a steeper angle - let's say 20-30°.

   Oh. I'd not thought of that at all. At those angles, the shroud design and
the size of the "window" the light sensor can see through get to be tougher
problems.

> From my laser sensor experience, I tend to think that
> retroreflective tape would work well...

   It may be because a good retroreflector won't work in this situation - the
light source (illumination LED) and phototransistor are offset, so any light
reflected right back to its source will not be seen by the phototransistor. If
the distance between the lightsensor and the target is large compared with the
distance between the light source and the phototransistor, there's no problem.
But with the light sensor only 2 cm off the surface, this is possibly not the
case.

> Rob's suggestion about ambient light shielding is
> very wise too...

   Yes, although as I mention, it was something I (thought!) I addressed early
on to get the error rate as low as it is. I really need to put together a
monolog of the barcode reader development like I've done for LIDAR or sumo
strategies.

-- 
Brian Davis