Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

wang siqi
Hi all,
    Does anyone apply the qgroundcontrol on the paparazzi uav successfully? If so, could you tell my the way ? (the imu calibration is too difficult for me to use, I have tried several times to calibrate the magnetometer, however, my quadrotor still has the tendency to rotate by itself when taking off...) Any help would be appreciated!
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

Hwarm
Hi,
sometimes the motor current wires  or  the battery
wires  are to near to the magnet sensor (keep it 6-10cm away).
If you use a switch wire the plus and minus wire to the switch.
Twist all wires.
Often it helps to use a double side PCB for current allocation.

Make this brutal test:
Hold the copter on the ground  and give  full  throttle to all motors.
Under this condition the magnetic direction  must have not more then
5°declination.
Sometimes it is a hard way to get good results.






Am 14.03.2016 09:38, schrieb wang siqi:

> Hi all,
>      Does anyone apply the qgroundcontrol on the paparazzi uav successfully?
> If so, could you tell my the way ? (the imu calibration is too difficult for
> me to use, I have tried several times to calibrate the magnetometer,
> however, my quadrotor still has the tendency to rotate by itself when taking
> off...) Any help would be appreciated!
>
>
>
> --
> View this message in context: http://lists.paparazziuav.org/Have-anyone-successfully-combine-paparazzi-uav-with-the-qgroundcontrol-tp17775.html
> Sent from the paparazzi-devel mailing list archive at Nabble.com.
>
> _______________________________________________
> Paparazzi-devel mailing list
> [hidden email]
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

flixr
Administrator
Hi,

You can load the mavlink module to use most basic functionalities of QGroundControl, but that does not help you in any way with your magnetometer calibration!

You should follow Prof. Warmers suggestions and do that test. If current is the problem, you can also try to calibrate for that:
https://wiki.paparazziuav.org/wiki/ImuCalibration#Calibrating_for_Current

I also started writing a simple mag calibration script that reads the IMU_MAG_RAW data live and directly displays the data in 3D while recording it:
https://github.com/flixr/paparazzi/blob/34e34e0b25d2fcf4c4e0c135079bbbb59a31f023/sw/tools/calibration/calib_mag_live.py
When you have enough points covering good portions of the sphere, close the plot window and it will continue and output the calibration.
Haven't had time to improve the script further yet..

Cheers, Felix

On Mon, Mar 14, 2016 at 7:49 PM, Prof. Dr.-Ing. Heinrich Warmers <[hidden email]> wrote:
Hi,
sometimes the motor current wires  or  the battery
wires  are to near to the magnet sensor (keep it 6-10cm away).
If you use a switch wire the plus and minus wire to the switch.
Twist all wires.
Often it helps to use a double side PCB for current allocation.

Make this brutal test:
Hold the copter on the ground  and give  full  throttle to all motors.
Under this condition the magnetic direction  must have not more then 5°declination.
Sometimes it is a hard way to get good results.







Am 14.03.2016 09:38, schrieb wang siqi:
Hi all,
     Does anyone apply the qgroundcontrol on the paparazzi uav successfully?
If so, could you tell my the way ? (the imu calibration is too difficult for
me to use, I have tried several times to calibrate the magnetometer,
however, my quadrotor still has the tendency to rotate by itself when taking
off...) Any help would be appreciated!



--
View this message in context: http://lists.paparazziuav.org/Have-anyone-successfully-combine-paparazzi-uav-with-the-qgroundcontrol-tp17775.html
Sent from the paparazzi-devel mailing list archive at Nabble.com.

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel



_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

Yannick Jestin

> On 14 Mar 2016, at 21:12, Felix Ruess <[hidden email]> wrote:
>
> Hi,
>
> You can load the mavlink module to use most basic functionalities of QGroundControl, but that does not help you in any way with your magnetometer calibration!
>
> You should follow Prof. Warmers suggestions and do that test. If current is the problem, you can also try to calibrate for that:
> https://wiki.paparazziuav.org/wiki/ImuCalibration#Calibrating_for_Current
>
> I also started writing a simple mag calibration script that reads the IMU_MAG_RAW data live and directly displays the data in 3D while recording it:
> https://github.com/flixr/paparazzi/blob/34e34e0b25d2fcf4c4e0c135079bbbb59a31f023/sw/tools/calibration/calib_mag_live.py
> When you have enough points covering good portions of the sphere, close the plot window and it will continue and output the calibration.
> Haven't had time to improve the script further yet..

   ENAC Student have written a "not so shabby" python/Qt program to perform all three calibrations, we hope to release them soon. If anyone is interested in cleaning up the code, feel free to jump in !


  Y.


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

flixr
Administrator
Hi Yannick,

cool! Did you also integrate a live view?
Looking forward to your pull request :-)

Cheers, Felix

On Mon, Mar 14, 2016 at 9:19 PM, Yannick Jestin <[hidden email]> wrote:

> On 14 Mar 2016, at 21:12, Felix Ruess <[hidden email]> wrote:
>
> Hi,
>
> You can load the mavlink module to use most basic functionalities of QGroundControl, but that does not help you in any way with your magnetometer calibration!
>
> You should follow Prof. Warmers suggestions and do that test. If current is the problem, you can also try to calibrate for that:
> https://wiki.paparazziuav.org/wiki/ImuCalibration#Calibrating_for_Current
>
> I also started writing a simple mag calibration script that reads the IMU_MAG_RAW data live and directly displays the data in 3D while recording it:
> https://github.com/flixr/paparazzi/blob/34e34e0b25d2fcf4c4e0c135079bbbb59a31f023/sw/tools/calibration/calib_mag_live.py
> When you have enough points covering good portions of the sphere, close the plot window and it will continue and output the calibration.
> Haven't had time to improve the script further yet..

   ENAC Student have written a "not so shabby" python/Qt program to perform all three calibrations, we hope to release them soon. If anyone is interested in cleaning up the code, feel free to jump in !


  Y.


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

wang siqi
In reply to this post by wang siqi
Hi all,
       Thank you for your reply, thanks a lot!
       Today I take the advice of Hwarm to keep my aspirin IMU away from the main current, just like the way we treat our GPS, my lisa m board is held high from the airframe.
       But that is not the point.
       The fact is, when I accidentally make my rotors rotate at the opposite direction to what the paparazziuav.org says, my quadrotor take off normally, without the tendency of rotating itself anymore. the specific address is http://wiki.paparazziuav.org/wiki/Rotorcraft_Configuration#Motor_Mixing
The direction that picture of time cross says is opposite to mine.
The direction of rotation of my time cross quadrotor is NW(CW),NE(CCW),SE(CW),SW(CCW),which seems to be the correct one.

      Anyone can explain about it?
      Thank you again!
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

flixr
Administrator
Hi,

can you please post the paparazzi version you are using (output of ./paparazzi_version) and the motor mixing configuration you are using?
Maybe even easiest put your whole airframe file on pasetbin or so...

Cheers, Felix

On Tue, Mar 15, 2016 at 12:20 PM, wang siqi <[hidden email]> wrote:
Hi all,
       Thank you for your reply, thanks a lot!
       Today I take the advice of Hwarm to keep my aspirin IMU away from the
main current, just like the way we treat our GPS, my lisa m board is held
high from the airframe.
       But that is not the point.
       The fact is, when I accidentally make my rotors rotate at the
opposite direction to what the paparazziuav.org says, my quadrotor take off
normally, without the tendency of rotating itself anymore. the specific
address is
http://wiki.paparazziuav.org/wiki/Rotorcraft_Configuration#Motor_Mixing
The direction that picture of time cross says is opposite to mine.
The direction of rotation of my time cross quadrotor is
NW(CW),NE(CCW),SE(CW),SW(CCW),which seems to be the correct one.

      Anyone can explain about it?
      Thank you again!



--
View this message in context: http://lists.paparazziuav.org/Have-anyone-successfully-combine-paparazzi-uav-with-the-qgroundcontrol-tp17775p17787.html
Sent from the paparazzi-devel mailing list archive at Nabble.com.

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel


_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

flixr
Administrator
Also if you are using a times cross quadrotor with NW(CW),NE(CCW),SE(CW),SW(CCW), then the motor directions are exactly the other way round than in the picture...
The you should use
<section name="MIXING" prefix="MOTOR_MIXING_">
  <define name="TYPE" value="QUAD_X"/>
</section>
instead of
<section name="MIXING" prefix="MOTOR_MIXING_">
  <define name="TYPE" value="QUAD_X_CCW"/>
</section>
Cheers, Felix

On Tue, Mar 15, 2016 at 9:04 PM, Felix Ruess <[hidden email]> wrote:
Hi,

can you please post the paparazzi version you are using (output of ./paparazzi_version) and the motor mixing configuration you are using?
Maybe even easiest put your whole airframe file on pasetbin or so...

Cheers, Felix

On Tue, Mar 15, 2016 at 12:20 PM, wang siqi <[hidden email]> wrote:
Hi all,
       Thank you for your reply, thanks a lot!
       Today I take the advice of Hwarm to keep my aspirin IMU away from the
main current, just like the way we treat our GPS, my lisa m board is held
high from the airframe.
       But that is not the point.
       The fact is, when I accidentally make my rotors rotate at the
opposite direction to what the paparazziuav.org says, my quadrotor take off
normally, without the tendency of rotating itself anymore. the specific
address is
http://wiki.paparazziuav.org/wiki/Rotorcraft_Configuration#Motor_Mixing
The direction that picture of time cross says is opposite to mine.
The direction of rotation of my time cross quadrotor is
NW(CW),NE(CCW),SE(CW),SW(CCW),which seems to be the correct one.

      Anyone can explain about it?
      Thank you again!



--
View this message in context: http://lists.paparazziuav.org/Have-anyone-successfully-combine-paparazzi-uav-with-the-qgroundcontrol-tp17775p17787.html
Sent from the paparazzi-devel mailing list archive at Nabble.com.

_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel



_______________________________________________
Paparazzi-devel mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
Reply | Threaded
Open this post in threaded view
|

Re: Have anyone successfully combine paparazzi uav with the qgroundcontrol ?

wang siqi
Hi flixr,
         Thank you for your help!
         You are right, my motor mixing configuration is
<section name="MIXING" prefix="MOTOR_MIXING_">
    <define name="TYPE" value="QUAD_X"/>
  </section>
Therefore the information on the website is correct, I have never noticed about this .
But I still have some questions, the biggest one is that why my quadrotor will float without the specific direction when taking off, you know, it can not still on the one point in the sky, and my mode is AP_MODE_ATTITUDE_RC_CLIMB. Is it about the problem of GPS or magnetometer or the barometer? could you give me some advice? Thanks a lot!