Chris,

You might consider using an old printer for its printhead control mechanism and attach a CCD camera head onto it and orient it for the validation test. I assume that the camera will be challenged to track a line on an unrolled paper pattern (note another usefulness in using the printer mechanism - you can feed the test pattern in through its roller mechanism.

My choice would be to use CCD direct but while I'd prefer control over the flash converter, that's probably not easily available in commercial modules.

Using photodiodes would be BORING. I assume the reason for this project is to demonstrate some of the basic engineering ideas behind making a road following system which you may have seen on some informative television shows. In that reference frame a camera approach would be more significant and parallel.

One advantage the camera would have over the photodiode is that due to its zone of view, it could better smooth the tracking when dashed lines must be tracked as are used in some road marking systems. When you enter the space between broken lines, the last thing you'd want to resort to is a back and forth SEEK TO ACQUIRE LINE pattern on a photodiode system; particularly if the idea is to suggest vehicle navigation possibilities.

What type of budget does this have? This affects the camera choices.

aka j