designing a quadrotor from head to the toe

[Phoenix]

Hi Dear Sonsivri Guys,
This is my first real post here and I'm so exited about being part of such a community, Now let's get straight to the business.

As a part of our bachelor project me and my friend decided to run a quadrotor project. so we bought the frame from http://www.rctoys.com/rc-products/K-DF-FRAME.html and decided to build the control board ourselves.
The IMU I chose was chr-6d and I have bought and additional magnetometer (we have not used this on the current board)
but I can't control it (
first of all,(this is a little embarrassing)  I'm not sure weather the direction of rotation of the blades, kindly watch this
http://www.youtube.com/watch?v=kvxHg82fpBg

Below is one of our test movies... as you see it flies randomly:
http://www.youtube.com/watch?v=DIP34T5SFtw

I am looking forward for your suggestions... thank you in advance

[th_sak]

Hi, First of all you sould know the principle of motors spinning. I attach an image to show you how you sould rotate the motors. I suggest you to search in IEEE and ScienceDirect for papers about quadrotors.

Gift paper from me!
Autonomous miniature flying robots Coming soon! - Research, development, and results
Link: http://ifile.it/w9rysaz

[Phoenix]

As you probably guess, I have read dozens of papers and thesis to reach here   never thought it would be so difficult.
but when it reaches to actually make one of them work,,, that's the hard part. I want to know if anyone has reached here and how does he get passed this stage. let me get you more insights... I have PID on pitch and roll and PI on yaw rate. my IMU outputs are pitch, roll, pitch rate, roll rate, yaw rate, accel on x,accel on Y and accel on z. But these outputs have errors and as long as a stable flight is concerned even these small errors count.
pid coefficients are guessed... based on trial and error and quadrotor is "stable" (so called   ) but when you let it loose, it just goes to where it wants.
BTW, thank you so much th_sak, Hope santa brings you sth nice too,
I'm sure your paper will help...
second BTW, where is my promised image? 

[LabVIEWguru]

Phoenix, you have GOT to see this site! (maybe you have already)

http://www.mikrokopter.de/ucwiki/en/MikroKopter?action=show&redirect=FrontPage

he gives the complete design, including code! I like the part of the video where he hovers it, then commands to "come home."
(I used to think I was a pretty smart boy, then I learned there are some REALLY smart people in this world!) I understand it's a hex-rotor, but there is probably much to learn from this.

update:
maybe you've seen them all, I never realized there was so much of this on youtube:

http://www.youtube.com/watch?v=nTeqQQ0V1Ro&feature=related

there are a LOT of quad rotor and copter videos - really, really cool! Good luck!

[th_sak]

second BTW, where is my promised image? 

I posted this image but I guess you can't see it because your are inactive user. Anyway I'll give you the direct link.
http://img843.imageshack.us/img843/7844/motprinc.png

You will find the answer in the papers you already read and in pages like this:

http://www.diydrones.com/profiles/blogs/arduimu-quadcopter-part-iii

Another gift: http://ifile.it/4px0bty

[Phoenix]

Did I mention you're a life saver? No I think forget that. But u truly are... thank you so much... about that photo, yeah
I've seen that before but my blades (apparent in the video) have a special curve and I'm not sure about that... CW or CCW is pretty obvious... but those papers you provide me with... awesome.... let me finish my reading materials and then I'll come back to you again

[LabVIEWguru]

That is AWESOME! (wait a minute - I used the word awesome?)  How long is the useful battery life? Is there a tradeoff between propeller size / speed / battery life? I ask because all the fliers I've seen use the small plastic propellers. I bet you can make all the cats move out of the area! This is brilliant work!

[Phoenix]

The real small motors u c in the videos consume about 10 A... yes ten AMPERS... I use 1700mAh 25c li-poly which work about 5-10 min and takes one hour to recharge
I'm not native speaker but my English is awesome   

[th_sak]

I am very interested in this topic because I'm about to start my Master thesis which will be based on FDI (Fault Detection and Isolation) of a quadrotor helicopter. I have many many many interesting papers and I also have experience in building and controlling such devises using MATLAB. To be more specific I helped some guy who was elaborating his final year thesis on my institution and we build together a custom-made aluminum frame with 4 brushless outrunner motors and 3 blade propellers. Draganflyer is good but has a lot of vibration in its frame and its motors have 1/4 power of those we used.

What we used:

Two of this:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5249

Two of this:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5247

Four of this:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5684

Four of this:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=2163

One of this (battery will last for 30 min flight at 80% of speed):
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=14370

One of this:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=6507

One of this:
http://store.diydrones.com/ArduIMU_V2_Flat_p/kt-arduimu-20.htm

For controller we designed a custom PCB powered by an Arduino (ATmega328).
For RF communication between MATLAB and Quadrotor we used two RFM12 transceiver modules:
http://www.futurlec.com/Radio-433MHZ.shtml

Total cost was about 200 euro.

Three more papers to read!!!

Development of a Micro Quad-Rotor UAV for Monitoring an Indoor Environment
http://ifile.it/9oh23p0

Mathematical Modeling and Nonlinear Control of VTOL Aerial Vehicles
http://ifile.it/lnhorzc

Step by step robust nonlinear PI for attitude stabilisation of a four-rotor mini-aircraft
http://ifile.it/btdl240

[Phoenix]

Interesting story about batteries... I use 3 cell (motors need 11.1) and also weight is critical for us... so I have to keep it less than 2000mAh. for communications I use Xbee which has been of great use, It never disappointed me. and my PC development platform is QT. matlab was not good at handling realtime problems ( In my opinion of course, maybe I was wrong) and my CPU is PIC18f6520
Could you actually make it hover? a stable fight? that's great... glad to meet someone so experienced... in order to tune your controller parameters... did you use a systematic model based approach or you used trial and error like us?
PS: my friend is always nagging about the vibrations of the frame, and I'm the one always telling him that it should work. ( Actually these vibrations are our biggest problem, you all have seen the foam we used right? )
As an Engineer, I'll find a way or I'll make a way...

[th_sak]

As I have already mentioned I was helping him and that was not my project. His quadrotor managed to fly and had the ability to lift up approximately 1Kg. Programming was made in Arduino language and as far as I know he used ArduIMU's library which has all necessary parameters to get the desired information from sensors. He didn't use any controller (PI, PID), he was just moving it by running on and off in various speeds the motors based on the data from IMU. Matlab was there to talk with RF module and get some data back from IMU just to plot the readings. It didn't perform any calculations. Finaly it didn't fly as we expected, but it did fly. I haven't started my thesis yet. I'm reading papers and try to figure out how to model the system with equations and then I'll try to implement various controll methods (PI, PID, H∞, Fuzzy, etc). Trial and error is not accepted.

[LabVIEWguru]

I hope perhaps these will be a help to you:

http://hotfile.com/dl/92148291/2419174/Unmanned_Aircraft_Systems.pdf.html

http://hotfile.com/dl/92149260/110fc12/4431538550FlyingRobots.pdf.html

Another book:

RC Model Aircraft Design. Not exactly what you are looking for, but maybe it has solutions to common problems:

http://hotfile.com/dl/88728302/49b45b0/RC0911295402mad.rar.html

http://www.filesonic.com/file/41329197/RC0911295402mad.rar

[Phoenix]

I didn't know there are books partly dedicated to quadrotor...

[LabVIEWguru]

Please post links to your progress. I think this would be extremely interesting. My eldest son has built several R/C airplanes. Recently he and some other friends have built "minimalist" airplanes by working with a type of construction foam. The wingspan was about 6 feet (2 meters) and was powered by a "pusher" propeller. He put a camera in the belly and recorded the flights until a wing cracked and it came down in pieces.

Why do your aircraft use such small propellers? Is it a matter of efficiency or speed of the motors or are you just buying what is available? I know the size and the pitch of the propeller or rotor is very important.

nidhisahu and pappinait1 love book as much as I - perhaps they have some books on autonomous flying vehicles.