Considering arduino for a photocell light installation, needing advice

I'm looking to make an art installation involving light. My searches for arduino projects that have similar concepts with my installation idea haven't been successful, so I'll ask straight.

I am wanting to make a photocell type project with arduino. The idea would involve the arduino controlling LED par can lighting. I would want the photocell to control the lighting entirely. I would want the photocell to have a set amount of range - for creating an example for discussion, say 15 feet. When someone is fifteen feet away from the photocell, the lights are entirely off. When they are 14 feet 11 inches away, the lights are dimly on. It would gradually increase until someone is a foot away or so, and the lighting is then 100% on.

I would also want the intensity of lighting to be matched with two voltage signals to a voltage-controlled synthesizer. The voltage amounts would need to be set ranges, so I could have control over the pitch and intensity the synthesizer is engaged. The idea being that the stronger the light, the more intensity the timbre and pitch of the synthesizer becomes.

My questions:

Would a photocell (presumably infrared, since we're dealing with needing absolute darkness for part of the installation) be able to handle this?
Would arduino be capable of coding this for DMX-managed lighting?
Would I need one arduino to handle both the photocell and control voltages?
And, ~most importantly~, what things would I be looking to buy for this project?

Once I get a good idea of where to go I'll be able to ask more concise questions, if I even need to ask more. Its just hard to get an idea of what I want to do when "photocell arduino" brings up sooooooo many scattered topics :frowning:

EDIT I am writing an artist's grant for this installation. I see that some topics here are based around "how do I do _______ on the cheap". While I still need to be in a budget, I have wiggle room!
EDIT #2 If there is a topic already about this that I wasn't able to find through searching, I'm more than fine with using my head and changing the project details to fit my own. Not needing someone to hold my hand through every detail.

Just thinking off the cuff, but the problem of a light sensor when you are turning on lights is you need to isolate the sensor from the lights being turned on. I've not used one before, but I would think an ultrasound ping sensor might be better for locating people coming to see your project: http://arduino.cc/en/Tutorial/Ping?from=Tutorial.UltrasoundSensor

Hallu:
Would a photocell (presumably infrared, since we’re dealing with needing absolute darkness for part of the installation) be able to handle this?

IR distance sensors are more suitable for ranges < 1m.
Above that, you’re better off with an ultra-sonic (“sonar”) sensor.

Liks this for example (13ft range):

Would arduino be capable of coding this for DMX-managed lighting?

Yes, you’ll need a DMX shield. Something like this maybe:
http://www.robotmesh.com/dmx-shield?gclid=CPrV8tbg9rkCFcwWMgodUDkArA

Would I need one arduino to handle both the photocell and control voltages?

Yes.

And, ~most importantly~, what things would I be looking to buy for this project?

I would probably buy:
1x Arduino UNO Rev.3
1x DMX Shield (link above)
1x Ultrasonic sensor (link above)

I think I’ve found a match!

I’m needing quite a long range (the longer and more dramatic the approach, the better), but the installation is in a quite confined walkway space (only about three feet or so of usable width). I need the sensor to notice someone approaching from a distance, but not so wide an angle that it also detects the edges of the hallway itself. Since most people will be walking in the center of this three foot walkway, I’ve found this sensor:

It has an eleven-inch wide detection sensor that works up to twenty feet in distance. Since people are roughly 1’6" wide, unless they were scrunched up towards a far side or purposefully turning sideways to avoid the sensor, they’d probably be triggering the sensor the duration of the walkway.

The specs seem to work for my concept, but does anyone think using this would be an issue for the installation?