Note: This is more of a proposal or idea than a set of instructions for how to build something. I thought "enable collaboration" meant "collaborative design of something that doesn't exist yet". But it seems Instructables are only for things that already exist.
Cities have a number of problems due to the huge numbers of people that frequent them. One is that irresponsible slobs (including lazy garbage collectors) leave trash all over the streets. Another problem is the excessive pigeon crap left everywhere because the excessive edible trash is conducive to them breeding all year round. How do we improve both problems at once? Train pigeons to pick up litter!
This was inspired by a great idea I saw on the halfbakery; I didn't come up with it. That's generally a speculative, sort of half-serious site, but I think this is one of the ideas that might actually work. I wonder if such a device could be designed and built collaboratively...
So my first mental image was big bins in places that are accessible to birds but not frequented by humans. They'll accept certain types of garbage, and dispense tasty pigeon treats in exchange. Teach a few pigeons to drop garbage in the bins and get a reward, and the rest should learn by example.
(I thought it might improve the poop situation by moving the pigeons to an area away from people while they eat, but I don't know if that would really work. Where do pigeons poop?)
One problem with this idea is that you would need to encourage them to pick up things that are actually trash, and not things like stones or hats. (How much can a flying pigeon lift?) The first thing I imagined is some kind of scanner like they have for bottles and can recyclers, but that's quite complicated and very error prone. How would you identify squashed, torn paper cups?
Although a lot of people hate them, pigeons are pretty smart. They can distinguish between Picassos and Monets, for instance, and were even considered for guiding missiles in World War II. Identifying pieces of trash should be easy; they already do while scavenging anyway.
So I wonder if there's some more elegant way...
Step 1: Cigarette Butts!
Cigarette butts are a universal problem ( cigarettelitter.org says that they are the most littered item in the world; in the billions per day, and they're carcinogenic and toxic to fish as well). They are uniform in shape and color, making it easy for a machine to recognize them pretty robustly. They are small and light, so no problem at all for birds to pick up. So now I'm imagining, for the first proof-of-concept implementation, a device that is meant exclusively for cigarette butts. This would greatly simplify the problem while still doing a lot of good (if it works).
Try to imagine your own version of how such a device would work before opening the next page, so your ideas aren't tainted by mine. :-) Leave your alternative good ideas in the comments.
Step 2: Passive Mechanical Device?
So then I imagined a device with a little hole on top for the birds to insert cigarette butts. It would suck them in with wheels or something, make sure they are actually butts (measuring the length and color?), and then dispense a piece of food.
The piece of food would have to be very nearby the deposit hole or other birds would learn to stand nearby and steal it when it pops out.
But even this is rather complex. Does it really need to be? False positives could just be sorted out after the fact just by scanning through the collection visually when it is periodically emptied. Hmmm...
So what about a purely mechanical, passive device? The bird has to physically push the cigarette butt with a little force, which pushes some type of mechanical linkage/lever, and then a similarly-sized pellet of food is propelled out the same hole, right near the bird's mouth. The food can be reloaded into the dispenser by gravity, with a hopper on top, and the butts can fall into a container below.
We'd need to worry about the thing jamming, I bet, and squashed butts might not have enough structural integrity to be "pushed" into a hole against a little resistance. It would probably need to be low-maintenance, but maybe not. Maybe getting jammed once a week wouldn't be that bad, if it's emptied once per day by a worker.
The container could not be a bag, and could not have gaps anywhere, or the birds would just take them out and cycle them through again.
But a rough sketch is not the same as actual instructions for building a real device. Please leave comments to help design this. What other practical difficulties would it have? How would the mechanics work? I'm not a mechanical engineer; I can't see this stuff intuitively.
Step 3: Training
I have no idea how you would go about training pigeons, but we know that they learn by example, so you should only need to train a few and the rest will follow. This is kind of how you train cats to use the toilet:
1. Dump some butts on the ground (or maybe "dummy butts" that have the same properties but are not harmful to the environment if they don't get picked up)
2. Feed some pigeons in the area around the device to gather them in a group.
3. Then feed them from around the device itself
4. Then put the fake butts in the machine and allow them to retrieve the food that comes out
5. Then put some fake butts on the machine itself?
Ok. I've explained what I am imagining for this device. Now everyone else fill in the gaps!