Quadrotor Build Thread

anwar · June 03, 2010, 04:54:59 PM

Ardu-IMU has an arduino processing core with an "Inertial Measuring Unit". This link has a list of what all functionality is included in it.

https://store.diydrones.com/ProductDetails.asp?ProductCode=KT-ArduIMU-20

There are two versions of it. The V2 one above has all the gyro + accelerometer stuff built in. Another V1 version lacks all the gyro stuff, and you have to get them separately and interface it.

http://store.diydrones.com/product_p/br-arduimu-01.htm

The software is open source in any case.

Gaurav · June 03, 2010, 05:51:44 PM

Arduino is a platform based on the atmel brand of micro-controllers. IMU - Inertial Measurement Unit consists of a collection of gyroscopes and accelerometers. The complete codebase has been provided as above by anwar

Gaurav · June 03, 2010, 07:39:51 PM

An Awsome guide to the basics of an IMU board.
http://www.starlino.com/imu_guide.html

sunk? · June 03, 2010, 09:06:02 PM

hi guys.
ok .......that WAY above my head there

so basicly................. is it a board you hook the motors and rx up to and it does its 'helicomand' type thing ?
regards
don

anwar · June 05, 2010, 07:33:11 PM

Quote from: anwar on June 01, 2010, 06:14:33 PM
Since our club is interested in a quad build, I have ordered this also : https://store.diydrones.com/ProductDetails.asp?ProductCode=KT-ArduIMU-20

Basically, this has everything one needs to build a quad.

I realized that there is one more component I will need (unless I hack up one of my receivers). Basically, the above board works on a single PPM signal from the receiver, as opposed to a bunch of individual PWM signals (one for each channel). So I will need a PWM to PPM converter (unless I can figure out how to tap PPM signal on a dirt cheap receiver, some people have tapped Corona receivers for this purpose).

With something like this : http://store.diydrones.com/product_p/br-ppme.htm , one can take 8 PWM channels (basically 8 servo outputs) from a reciever, and encode them all into a single PPM signal. So only one input pin is needed on the arduino side.

If anyone is going with diydrones.com stuff and minimal DIY, it is better to wait for their ArduPilot Mega product, which seems to have this also built in. It is supposed to be just connections and no programming. On the flip side, it has way too much functionality, and not sure about the price either.

anwar · June 05, 2010, 09:04:58 PM

ArduPilot Mega is receiver ready :

http://diydrones.com/profiles/blogs/ardupilot-mega-home-page?xg_source=activity

rcforall · June 05, 2010, 10:28:00 PM

Anwar ,
Looks like the codes will be released around early July from Chris's replies

anwar · June 27, 2010, 02:33:22 PM

Got the equipment to try both setups (using better vtail mixers, and using the complicated microcontroller based version) !

Tests using the vtail mixer based one is simple, just replace the mixer, tweak and test.

But the one using the microcontroller involves some delicate soldering and playing with software. Haven't done ISR level C coding since degree project involving decoding viruses in assembly and writing vaccines for them in C the early 90s (wonder if anyone remembers the MichaelAngelo virus

), so this would be a good journey back in time. The only concern is time !

izmile · June 27, 2010, 04:08:29 PM

I used to have a collection with full source of those DOS era viruses.. Must be somewhere in my study back in India.

At that time, I was particularly amazed by the virus "Happy birthday Joshi"... whoever wrote that had a true command of the computer architecture.. Was quite a learning for me and picked up some neat way of performing multiplication/divisions of large number.

anwar · June 27, 2010, 04:43:05 PM

Ah, I was hoping/expecting someone who had interest in those areas (and from that era) to respond

I even got the LBBV (http://vxheavens.com/lib/vml00.html) ordered from the US through a friend, to see more samples of code (back in 91!).

BTW, the board I ordered (with 3 axis gyro and 3 axis accelerometer) comes with code, but original design has an extra magnetometer and related code too. And I will be forced to tinker with the code to remove the magnetometer code, and see what additional stability can be derived using the other sensors.

izmile · June 27, 2010, 07:47:33 PM

If you have the interest and time, try the Giant black book by the same author. He covers most of the security threats including modern day worms.

anwar · June 28, 2010, 12:31:33 AM

Nope, moved on to other things (business apps, scalability etc)

But those days were fun. Just using "DEBUG" that comes with DOS, it is fun to disassemble the virus code, and then writing a vaccine in C (including neutralizing it in memory first... the magic of "cli()" and "sti()").

Now this Quad project should be fun. I had not come across the use of a magnetometer for stabilization until a few weeks ago. That was a whole new dimension to the problem.

anwar · July 10, 2010, 01:55:58 AM

Quote from: anwar on June 27, 2010, 02:33:22 PM
Got the equipment to try both setups (using better vtail mixers, and using the complicated microcontroller based version) !

Tried one more brand of mixer, did not have much success. In fact, it messed up the outputs so bad, that just getting the ESCs to arm became pretty much impossible.

So moving on to the microcontroller based design. Got the ArduIMU+ and magnetometer soldered up. That itself was interesting. I hadn't done any precision soldering for a long time (over 15 years), so had to get everything fresh (low wattage iron with fine tip, thin solder, wick, etc). Downloaded the Arduino software development environment, and the code that is specific to this board. Changed some settings, and uploaded the code to the board. Then used their testing software that would dump out the values from various sensors, and show an airplane that moves as one tilts or moves the sensor board.

Here is a video of that testing.

ujjwaana · July 10, 2010, 02:08:34 AM

Wow!! Way to go for myself!! just as Vinay was telling that flying might become boring after a while, I guess there are lots of things to do!!

keep up posted Mr Moderator!! this looks pretty exiting!!

saurabhhsrivastavaa · July 10, 2010, 09:46:01 AM

Nice work Anwar !!!

Keep us posted. BTW, what is this magnetometer used for ?

Cheers!!!
Saurabh.

anwar · July 10, 2010, 10:36:17 AM

I am also quite new to magnetometers, especially in such applications. Usually in UAV type of applications, a GPS is used to figure out if the aircraft has drifted away from where it was before. But for short distances and indoor applications, this does not work, and that is where the magnetometer comes in. It helps determine positional drift in the yaw axis.

Sreedhar · July 10, 2010, 11:21:33 AM

gud work anwar bhai....

anwar · July 11, 2010, 01:24:35 AM

Quote from: saurabhhsrivastavaa on July 10, 2010, 09:46:01 AM
Keep us posted. BTW, what is this magnetometer used for ?

Here is a nice write up on how gyros, accelerometers and magnetometers all come together (and GPS too).

http://diydrones.com/profiles/blog/show?id=705844:BlogPost:41145

spitfire · November 06, 2010, 10:07:00 PM

Anwar, any update on your quadrotor....?

vulture · November 07, 2010, 07:03:27 PM

all,

i started a thread in the forum called quad rotor.. in that many asked me for the video .. just before i made a setup for a student with my controller .. pls comment on my video .. first flight of my controller with out any tuning .. she will be rock solid in the air with fine tuning ..

happy diwali

vulture · November 07, 2010, 07:15:13 PM

im making 8 rotors .. with dslr ... for aerial photography ... videos soon

anwar · November 07, 2010, 07:46:17 PM

Quote from: spitfire on November 06, 2010, 10:07:00 PM
Anwar, any update on your quadrotor....?

Haven't worked on it since I went on some long trips.

And after coming back, all RC/free time these days is spent on the forum, and training others.

In fact there are a bunch of things to be built, including a Trex 700 heli !

spitfire · November 07, 2010, 07:56:22 PM

Quote from: anwar on November 07, 2010, 07:46:17 PM
Haven't worked on it since I went on some long trips.

Did you find anybody who manage to fly it without using microcontroller board and just by using mixers and gyros...?

anwar · November 07, 2010, 08:05:28 PM

My research led me to two options.

1. There is a way you can build a mixer that can handle signals that arrive out of sync (as is the case here, where there are gyros feeding into mixers). But I could not find the code for it online, and did not have the time to work on it at the code level. Might try it someday, it is in the back-burner now. I have posted some RCG threads about this earlier in this same topic.

2. People say a mixer called "Digimix III" will work, but they are expensive. Might as well get a mC based one, which is what I did.

The diydrones.com guys now have a new board that is ready for quads out of the box. It came out after I got my stuff. With that, there is no need for separate PPM encoder etc, it is just an Arduino board and a sensor board tightly couple as an integrated unit.

chetankolhe · November 30, 2010, 01:31:52 PM

hiii
anwer
i was searching quad rotor helicopter procedure from last 3 months. i want to make my own
after i saw your thread finally i believe that i can make it and i am starting from today
so can u please help me out for throwout the project?
because i am beginner to make any kind of robot i think i have to face many problems
but i think u can help me out.