Go Down

Topic: Help save the Vultures with a 10DOF (Read 21147 times) previous topic - next topic

adambloch

Hi all

I have started a little project that in my nativity I thought would be a simple project to complete.

Let me give you a quick rundown of what why and how.

99.9% of the world's Asian Vulture population has been wiped out mostly by accident over the last 10 to 15years. I am involved with helping the captive breeding projects that are trying to save the Vultures. We are captive breeding to release Vultures not just back into India and the worse effected areas but also into Europe where there has been other problems that have led to the demise of these very important species. We also work on many other species such as Eagles, Owls, Kites and in fact almost all other birds of prey need some help.

In this modern era everyone seems to think we know everything about everything, but as many of you know this just isn't true. One of the problems of incubating eggs of Birds of Prey is that we just don't have the data of what happens under the birds when they are sitting on the eggs. Chicken yes Birds of prey no.

What I am trying to do is create an egg to monitor that happens to the eggs under the brooding birds. What I need this egg to do is basically what a 10DOF would give out plus I need to add an array of DS18b20 to the surface of the egg to measure the temperature gradient across the surface of the egg. I need this to transmit the data wirelessly and the batteries need to last 70 days. I would also like to data logged internally so that if data is lost via wireless the data can be recovered at the end. The live data needs to end up in a database but a graphical representation in real time or play back would be good (sorry just throwing it all in)

Now due to size constraints I have bought Microduino boards that seem to work (test scripts) but what I have realised is that I am fighting a fight here that I'm not sure by myself I will ever win. I'm open to suggestions on hardware as there maybe better out there that I haven't seen.

If you are willing to help someone who is trying anything and everything to help the Vultures and other species I would love to hear from you. If you don't care about the vultures but think you could help I would equally love to hear from you as well

Many thanks

Adam

PS I forgot I need to measure humidity as well :)

Coding Badly


Does everything have to fit inside the egg?

adambloch

#2
Jun 17, 2014, 11:53 am Last Edit: Jun 17, 2014, 12:05 pm by adambloch Reason: 1
Hi yes the whole idea of this is that we will substitue the real egg for the monitor one.

I have attached some pictures. The small egg is the smallest egg the I think we might be able to squeeze everything in to and the larger is a Vulture egg. Loads of room for batteries :)

The small stack is the Microduino modules all stacked together. The uno is for scale.

1 x Microduino-Core+
1 x Microduino-FT232R
1 x  Microduino-nRF24
1 x Microduino-SD
1 x Microduino-10DOF
1 x Microduino-BM

http://www.microduino.cc/wiki/index.php?title=Main_Page







adambloch

This is just to show you what comes out of one of these eggs :)

wwbrown

That is a face that even the mother couldn't love.

Even with the above comment I am willing to pitch in on this project.  I am an engineer by training so I try to frame any project early on with the requirements.  I understand that the requirements first route is kind of boring because we all would like to do something because that is exciting and fun, no chance of letting the magic smoke out when you are just working with pencil and paper.  Here are some questions I have after thinking about this project for 5 minutes:

  • Does the egg have to weigh roughly the same as other eggs to spoof the Vulture?

  • Can the egg be replaced if the battery life is a problem, you put the original in why not a second one?

  • Given the real eggs and the fake egg will have different weight distribution the fake will move differently from the real eggs, is this error going to be too high for the study?  One being liquid changing to a bird with structure makes it difficult to make a fake egg close to a real egg.

  • What is the maximum distance allowed between the egg and the RF recorder or repeater?


That sums up my concerns after giving the project a very small amount of thought, if you would like to have me help out leave me a PM.

Thanks
wade

adambloch

Hi Wade

I don't know what you mean she's beautiful :)

The weight of a Vulture egg is around 250grams but the birds don't ever pick them up so as long as it's not stupidly heavy I'm sure it would be fine. We use all sorts of materials form wood to plaster of Paris for them normally.

It would be much better if we don't have to replace the batteries and in the case of using this in the field that would be very difficult. The good news is that the most important data will come from the first 20days.

Weight distribution is something I have been thinking about but a real egg also isn't perfectly distributed they tend to be a bit lopsided, this due to the liquid and the chick inside. They don't just slop around keeping the egg balanced. I think that we will find that the bird if not turning the egg over fully in any case it's more 90 to 180 one way then the same back the other way but I might be wrong which is why we need to do this…

As far as distance goes.. In the captive birds we could put something right under the nest in the wild I should think we could be within 10m. The further the better as wires are a nightmare and vultures really don't like
strange objects..

Many thanks

Adam

Coding Badly


Is transmitting data necessary?

Would a "relay station" be possible?  The egg would transmit to something nearby that would have a much bigger power supply which would retransmit the data to a central collection.

What do you not like about the Microduino hardware?

The electronics can potentially be shrunk to an unbelievably small size.  I suspect the antenna and power supply are going to be your biggest obstacles.  If you can, please put two double-A batteries near / in the example egg.

adambloch

There's nothing wrong with the Microduino as far as I can tell. I just always belive that there maybe a better or different way of doing everything.

Yes relay station would be possible and the real time data is for two reasons

1. we then know that we are getting data and that we aren't just wasting time. If there is a fault and we need to swop the egg we would make that call.

2. It would be very cool :) I know on that face of it that sounds like a stupid answer but Vultures have a bad rep and anything that helps people be interetsed in the very important work we are doing is of great help. Also it would be nice to be able to have a real time video feed that we can look at the position of the egg on the screen as the bird moves it. we could then record both at the same time so that we have even more data on what actions and movements the bird does and the effect on the egg. We already stream from cameras on the nests and a simple PIP would be simple enough.

here are the pictures








Coding Badly


You have pictures of two dramatically different sized eggs.  Are the eggs from different species?

adambloch

Hi yes.
The small one is a yellow billed kite and is about the smallest egg size that I thought would be possible to use. But the more important one is the vulture egg. If Li-po s were used I think we could squeeze it in to the kite size as the smaller birds incubate quicker so battery life can be shorter.

A

jbarchuk

Excellent project. I've read about how the loss of the VULTURE population has resulted in the massive growth of the FERAL DOG population, which has led to an increase in human attacks/deaths.

I'd be very surprised if the commercial chicken/duck/goose breeder industry hasn't already created this device. Are you sure it's not available? Maybe other wild bird researchers? The CHICKEN industry in the US is a STAPLE COMMODITY. They spend $millions to increment productivity and this device -can't- be a new idea.

adambloch

#11
Jun 18, 2014, 05:31 pm Last Edit: Jun 18, 2014, 08:29 pm by adambloch Reason: 1
Not seen one yet... You are right they spend Millions on the chickens as they can get millons back out. You can get temp and humidity loggers that look like egg but I bet most of the research with the chicken is done by watching them up close. stick some paint dots on the egg and put a sheet of glass on the bottom of the nest :) No we can't do that with a vulture :)

Chicken egg are also very easy to hatch and the safe error level seems to be greater. Being precocial they are very tough unlike the chicks we have to deal with. The adult vultures in many cases have to help the chick out of the eggs. most of the data out there is based on chicken and this is another reason why I want to do the research.




luisilva

Very nice project!
The subject talks about 10DOF. What are the variables that you need to sense? (DOF means degree of freedom, right?)
In the subject of batteries there are big ass button cell batteries too, I believe that it can be a option too. But at the end, the best choice, I think that may be LiPo.
Why to you use "pre-made" boards, and not design you own board? It can have whatever you need, and can be of the size (or shape) you want. The other important point about this question is it can have ONLY what you need (and nothing more consuming your precious power).
Doing some math: if you have one battery with 1000mAh and you need it to work 70 days, the system need to consume at max 1000mAh/(70*24h) = 595uA. This must include the bursts of RF, save data in the flash/SD card etc. I think it can be done, but maybe not with any microcontroller/system.
The last question (that may be the key question) is about rate. At what rate you want to read data?


adambloch

#13
Jun 19, 2014, 08:27 am Last Edit: Jun 19, 2014, 10:16 am by adambloch Reason: 1
Thanks for the interest in the project. The data rate needs to be high enough that it's useful and we can see what is going on but I do see that power is going to be the killer here. The other thing that is being over looked is the surface temperature data. The data needed is to show what happens to the egg and how it is rotated by the parents but also the effect to the temperature gradient applied to the egg of the incubation by the birds sitting or not sitting on the on egg.  This is where I was looking at using the DS18b20s in the shell. Using a 3d printed case that has the holes of the DS18b20s already in it.

One thing that I hadn't thought about is the heating of the egg by the processor etc. I have been running the stack for 24hrs and there doesn't seem to be much heat from it bit I should really put it in a box and see what happens then. It needs not to get warm as the birds will stop sitting on it if they feel the egg is too hot.

So everything that the 10DOF can show plus the temp gradient. I.e Rotation, movement, acceleration, heading, temp etc.




luisilva

#14
Jun 19, 2014, 05:24 pm Last Edit: Jun 19, 2014, 06:02 pm by luisilva Reason: 1
One thing that I've noticed in your last picture, that helps me in my point of view, is that you have 4 LED's in your PCB stack! What they are doing if the stack will be inside one egg and no one can see it?! If they are "eating" 0,5mA each the 4 will "eat" 2mA. That is in my last math ONLY the led's will discharge the 1000mAh battery in only 500h (21 days). I suggest (at least) removing those led's or his series resistors from the pcb's.

Go Up