The Quad Rotor Observer flight by Jean-Louis Naudin

Hello to All,

I have done successfully the first test flight of the Quad Rotor Observer (QRO) v4 with a new costless flight controller board V7 which uses a Atmega 168 and 3 piezo gyroscopes (NEC CG-L43). This is a personnal variant of on the KKmulticopter board widely spread on the web. I am really surprised by the stability and the flight enveloppe of this quadcopter Vs the simplicity of the firmware code (Occam's razor principle...) and the low cost of the electronic material used here in this design...

Tested setup:
- personnal flight controller v7 with Atmega168 and 3 piezo gyro CG-L43 based on the KKmulticopter board
- firmware Quadcontroller v4.5 by Rolf R Bakke
- 4 brushless motors DualSky XM2822CA 1450KV 7A
- Lipo battery 3S (11.1V) 1500 mAh
- receiver Corona CR6D 2.4 Ghz

More technical infos at: http://diydrones.com/profile/JeanLouisNaudin

Comments

Ashik imran March 18, 2014 at 9:29pm

hi,Jean-Louis,

sir i very happy writing to you because without you i could not have any knowledge about quadcopter. sir i have only one small request, may i get C or another language programe code for the KKmulticopter circuit you provided. i will be very thankfull....thankfull....thankfull....please help me.
Sébastien Pruvost April 28, 2012 at 4:21am

Hello,

I want to build the KKcopter completely. That's why, I would like some informations about the KK flight controller board.
What do you think about the electronic diagram ? Is it equivalent to a real KK board ? Is it also operational than a real KK board ?
Best regards.
Mayank Gupta April 13, 2011 at 8:17am

I request you my dear friend .can you please help me to read the PWM signals from the receiver and write them through the microcontroller to the ESC

my website: http://www.quadrotor.co.nr
Vincent Mees March 8, 2011 at 10:30am

What is your landing gear made of?
Rana March 8, 2011 at 9:12am

From where you got CG-L43 ?
Developer

Jean-Louis Naudin January 5, 2011 at 10:42am

Hello Jason,

If you see the firmware code, when the throttle is set to 0 (<5%) the gyro are recalibrated... So, each landing the gyro are re-calibrated... see below:

ma82:

;--- Calibrate gyro when collective below 1% ---

lds t,FlagCollectiveZero
tst t
brne ma12     ;collective below 1% ?
rjmp ma4

ma12: lds t,CalibrationDelayCounter   ;yes, increase delay counter
inc t
breq ma50     ;delay reached 256?

sts CalibrationDelayCounter,t   ;no, skip calibration.
rjmp ma51

ma50: b16clr GyroZeroRoll    ;yes, set gyro zero value (average of 16 readings)
b16clr GyroZeroPitch
b16clr GyroZeroYaw

ldi counterl,16

ma20: rcall ReadGyros

b16add GyroZeroRoll, GyroRoll
b16add GyroZeroPitch, GyroPitch
b16add GyroZeroYaw, GyroYaw

Regards, Jean-Louis
Developer

Jason Short January 5, 2011 at 10:09am

Well, you'd be the first to try it. Anyway, wonderful as usual.

You could also include a reset switch to zero out the drift in flight... I did something similar using a training switch.

Jason
Developer

Jean-Louis Naudin January 5, 2011 at 7:02am

Hello, I have just conducted a Power/Weight ratio measurement with a DPR-50 datalogger on board of the QRO v4 during a stable hoovering at 2 meters above the ground.

The Take Off weight (with the DPR) was 668 g

The Average Power required for the hovering was 95.29 Watts

So, this gives a Power/Weight ratio = 143 Watts/Kg

Regards, Jean-Louis
Developer

Jean-Louis Naudin January 4, 2011 at 11:08pm

Hello Jason, Yes, why not... but at this moment the firmware works very well as you may see in the video of the flights and the drift time is greater than the flight time (about 5 min) ;-)

Occam's Razor principle can be applied here...
Developer

Jason Short January 4, 2011 at 10:44pm

Why not use the feedback from the user to correct for errors? If the user is sending commands to right the ship, you could use those to slowly correct for drift. If the gyro drift is not too bad that is. Maybe limit the integration to commands of 5 deg or less.

of 2