Ref: Laser Metrology

I was thinking the same thing with the encoders, I want to try out servos but
for me the price of encoders just to give something a try doesn't make sense.
I'm sure a cheap digital vernier could be hacked to get a signal,
great...until the coolant is turned on :)
Infra red could be an option but off the shelf rangefinders are nowhere near
accurate enough so it wouldn't be simple. A breaking beam encoder (like a
mouse with a ball) could do a good job with a dedicated microcontroller, 1
million steps a second would be fairly simple to get, 5 million may be
possible and even higher with (expensive) radio frequency PIC's but it would
still suffer from the 'where am I?' problem steppers motors have.
The webcam and laser pointer method could work well if the beam could be split
up and all the separate beams at a very steep angle. It may also be possible
to use a flat beam angled to go from the top to bottom of the image over the
length of the axis and at a very slight angle from right to left so as the
beam goes up or down the left to right pixels could be treated as a kind of
vernier scale.
Not sure if a CD laser would be powerfull enough in open air, I think some of
then are blue though ;)
cheers