Let's Make Robots!

Robotic Bee Colony

 

I was out for the weekend and it seems there was a lot of talk on this subject. It happened that I had one of my bee hives swarm and I had trouble retrieving it. However, this gave me a thought of changing the theme a bit: Robotic Bee Colony. 

There are 3 types of bees, the Queen that lays eggs, the Drones (male bees) that do nothing, only mate with virgin Queens and the Worker bees that do everything in the hive and in the field. I guess we can eliminate the Drones, or retrofit them with some dignity and let them do something useful. To simulate the life and work of a real bee colony, the RBC can be structured as follows:

There can be hexagonal cells scathered on the game field. The Worker bees can go out and gather them. They can have different colors inside, simulating pollen and honey. When the Workers get a cell to the hive, other Workers can use them to make comb by stacking cells and arange them in rows. There will be some empty cells where the Queen can "lay eggs" to make more Workers. When there is an empty cell surrounded by 3 cells of honey and 2 cells of pollen, an egg can be layed, and after a while another Worker bee will be "born" in the colony. To simulate the man's intervention extracting the honey and the pollen, either automatically or manually, the filled cells can be removed from the comb and scathered back in the field again.

The Queen should be more like the brain of the colony, it should send the workers to gather food, direct them to construct the comb, command them to defend the hive. Her Majesty will look for empty cells surrounded by honey and pollen to lay eggs to make more Workers. The eggs can be some small coin batery operated devices that have a counter that will trigger an IR signal to open a hatch from where a new Worker bee will enter the hive. 

The Worker bees will have an age counter. Freshly came out from the hatchery, they will have to search for the Queen to give them an ID number to know to which hive they belong to. Then they will follow the Queen's orders to make comb from the empty cells laying around and the honey and pollen cells the older Worker bees gather from the field. After they make some comb, they will became older bees and ask the Queen if they can go out to gather honey and pollen. The Queen allways needs at least one Worker bee that builds comb around her, but every new Worker has to build at least one comb cell (an empty cell surrounded by 3 honey cells and 2 pollen cell). After the Workers bring in at least 3 honey cells and 2 pollen cells they will became too old to "fly" and stay to defend the hive until their internal clock will make them "retire" to the hatchery to recharge. The Workers will scoot around to find honey and pollen cells, but they may end up at the entrance of another hive, if they can get in, they can steal the cells from there. So the guardian workers have to ask them the "password" to let them enter. If they can "steal" the password, they can enter the hive, if not, they will be pushed away.

This idea can be further polished and adapted, let me know what you guys think.

 

Update Sept. 2nd, 2012:

Las week I was thinking about this project idea again. I really want to do it over the winter. How much it will be done, I have no idea yet. So far I think the idea presented above is a bit complicated, so I decided to simplify it. I need to simplify the bee robots to be cheap enough to build many. So I guess that the hexagonal cells for the robots to assemble to make a comb will be scrapped. I'm thinking to use a row of 5 LEDs in the abdomen to show off the nectar gathering-exchange. In the hive, I think I'll just have 2 frames painted on the board to simulate a vertical hive like we actually use in real life. The top frame will be for honey deposit and the bottom frame will be for pollen on the sides and brood in the middle. LEDs will be used to show the honey, pollen and brood in the frames, installed underneath the board. I will try to simulate the comb making, so far I don't know how, but I'll think of something. Oh well, I'll keep you posted when I'll have more ideas.

 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Any new development here? Please update ;-)))

Nope, no development in any of my robotic projects, I've been busy with real life problems. Lots of work, work for the real bees plus I'm in the process to buy a condo-townhouse. This weekend I'll be at the farm working on bee equipment, so again, no robotics.

I did receive in the mail the 6 Bluetooth modules I ordered, so the first progress will be on the Football robots.

Ok, hope to see your football bots running soon.

I don't think these guys exhibit the swarm architecture that you're talking about, but they certainly remind me of bees in a way that would freak out Chris Farley:

http://techcrunch.com/2012/02/01/swarming-robots-will-fly-menacingly-towards-your-loved-ones-in-perfect-formation/

Oh yeah, they do look menacing! I would love to build a robot bee that actually flies and walks, but hey, that's over the budget.

Interesting information. I will keep that in mind when creating our swarm/hive world.

 

I was sick in bed all day so I watched the Ted talks on robots today. This guy's work reminds me of this thread. http://www.ted.com/speakers/hod_lipson.html UPDATE: I screwed up. It was this guy: http://blog.ted.com/2007/11/27/robert_full/ I watched a lot of robot TED talks today.

Thanks guys. I am not going to pursue this idea at the moment, though if I will build a colony of robots it will be based on this idea, because I am a beekeeper and I would like to showcase their life instead of ants.

To me it is important to mimic their real life as much as possible because at some point I can take the setup and explain to kids how bees work. The bees have a natural parasite that jumps on their back and sucks their blood, called Varroa. That can be an energy sucker they find in the field. Other bees do not do that. But the honey and pollen are food or like energy stores for them. That can be simulated somehow. The only thing that I can't really simulate is the growing of the bees inside a comb cell. It can work of a cell is about the size of the bee, bit then it will be hard for them fo carry the cells from the field to simulate honey and pollen gathering. Another thought might be to collect some yellow pellets as honey and red and blue pellets as pollen and actually store them in cells. The eggs can be white pellets. Thinking more about this, perhaps the cells can be readily made as comb with samll walls and the bees need to be able to go on top of it to drop the pellets. The bottom of the cells would be hinged to let the pellets fall through and let a bee come out. We'll see.

The bees have a short life span, about 28 days. They are doing several kinds of work inside the hive, feeding larva, feed the Queen, keeping warm, making comb, vent the nectar to evaporate the water, cleaning, guard the entrance, then, after about 18-20 days, they start flying to get honey, pollen and propolis. The flying wears them down quickly and they die. New bees are born after 21 days from the eggs layed by the Queen. The Queen can lay 2000 eggs per day if there is space and is properly fed by the bees in her suite and she will live as long as she can lay eggs. When she gets too old, she lays more Drone eggs (unfertilized eggs) and the bees will change her. They will grow a large cell and make the Queen lay an egg there, then they will feed it abundantly and with no pollen, only with royal gelly, and after 16 days a new Queen will emerge and she will kill the old Queen. After a week of getting fed honey, in a sunny and less windy day, she will fly to mate with the Drones. She will fly as many times as needed until she feels she has her sperm bag full, then she will not come out from the hive again unless she is swarming. Swarming happens when the hive is full of honey, brood and there is not enough space to house all the bees that form the colony. The beekeeper needs to be attentive and remove the honey and give the bees more space (add more boxes with frames) to expand their colony. When a colony gets too big, the beekeeper will split that colony in two and let one half to raise a new Queen, or he gives them a readily mated Queen. The Drones do not do any work inside or outside of the hive, their single purpose in life is to mate the Queen. Accidentally they keep warmth to the brood inside the hive, so the bees let them hang around just in case they need a new Queen. In the Fall, the bees kill all Drones and do not let them enter the hive anymore. No need for a parasite during the winter when every drop of food counts. Because they don't fly to gather nectar and pollen over the winter, they don't wear down and they live a few months.

There are many other aspects of a bee colony, but these are the most important. I am trying to showcase their life and work in a way that will be obvious to the viewer and this project may be an interesting educational tool. Now that would be fun.

 

Keep the good work up!!

A idea struck my mind regarding Varroa- You can use large value resistors to simulate those. The increase in the resistance of the circuit will lead faster discharge of batteries and that can simulate the lack of energy. Then they can go to a recharging station which will be full of 'pelletes' of honey and polen to recharge themselves. The recharging station can be a hexagonal block with solar cell on one side.

As to the drones, you can bring them near the white egg pelletes and the fertilisation can be actually bootloading the IC (Atmega 8 in my case).

And on the topic of the "Assassination of the Drones", I think bees would use thier stings for that. Am I right?!? If yes, then a small drill which actually is a drill bit mounted on a motor should work just fine.

I hope you ponder on these suggestions. Keep on brainstorming and maybe we'll have to run for our lives from a robo bee infestation in LMR!! Lol, only joking!! 

I think the batteries will discharge faster if there is a small resistance in the circuit, thus letting a large current to flow to the ground. But I do not intend to do that. The energy transfer is hard to see, it takes some explaining. Since the robots can run around for about 2 hours until they need a recharge, I thought that may be their life span. As soon as they will detect low voltage, they will go back to the hive and cease function over there. The cleaner bees will drag the dead bees out of the hive into a hole that leads to the hatchery where the dead bees will recharge and will get their "memory" erased so they can start a new bee life after recharging.

There are several aspects that I am still thinking of.

In real life the bees get nectar from flowers, transport it to the hive, give it to other bees that will mix it with an enzyme and deposit it in a cell. Then other bees will vent those cells with their wings to evaporate the water. After a while, the bees will get the thickened nectar out, mix it again with their enzyme and finally put it in the final cell where it will be vented again. When the cell is almost full, the bees will start capping the cell and leave only a tiny hole to fill it up completely, then cap it off. I would like to showcase this process, however, I don't know how to show the capping of the cells. 

Another aspect is the transformation from the egg to larva, then to capped brood and then the emerging of a new bee. The Queen lays the egg as a small white stick, then the bees feed it with royal jelly and after 3 days a larva appears from the egg, then the bees feed it with royal jelly the first 2 days and after that they add small portions of pollen. The larva grows until the cell is full (6 days), the bees cap the cell with a combination of wax and pollen to let the larva breathe. The larva will transform into a pupa (pupal stage takes 12 days) and then to a bee, the whole process takes 21 days. I would like to showcase this process somehow, perhaps with a light underneath the cell and some shadow that grows bigger as it is fed by the bees, then somehow cap the cell and let it in the dark. Then the cap is removed and a new bee emerges into the hive. So here are my thoughts: Every day will be a minute. A timer starts as soon as the Queen layed the egg. After 3 minutes, a roll underneath the cell rotates one position that shows a small larva, then after 2 minutes rotates again to show a larger larva, then after another 2 minutes rotates again to show a big larva that fills the cell completely, then after 2 more minutes rotates again to cap the cell, then after 12 minutes rotates again to let a bee robot come out. This might work...

Now let's talk about the actual robot bees. Of course, they will have a micro, motors, wings that vibrate, legs, antenae... I am trying to decide if I should use a couple of pager motors mounted as the middle legs and have fake front and back legs or actually make a hexapod robot (the simple one, with 3 servos). The chassys will be a PCB shaped as a bee. I'll use a AAA Li-Ion battery mounted underneath the abdomen together with the stinger with the micro on top of the abdomen. On the head, there will be 2 IR sensors and 3 small IR LEDs in the middle, exactly like the bees eyes. 2 antenae connected to switches will make bumper sensors. Underneath the head there will be the mandibles to handle the food and a color sensor to detect the type of food. Oh, on the head there will be colorful LEDs to show message communications between the bees and the Queen. The Queen will have no mandibles, but some kind of storage for eggs and a ejecting mechanism at the tip of her abdomen. The Drones will have just a bigger size and no extra function than walking. Perhaps I'll try to simulate mating too, but I'm not betting on that yet.