paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] Yaw from AHRS


From: Christophe De Wagter
Subject: Re: [Paparazzi-devel] Yaw from AHRS
Date: Fri, 28 Oct 2011 12:23:59 +0200

Dear Loic,

this looks like a magnetometer or sign issue. 

1) the gyro's are integrated. this is fast. (initial transient)  (and is done at the PROPAGATION_FREQUENCY which influences CPU load and very agile motions but should not influence normal operation)
2) then the magnetic measurement compensates for integration errors. This is slow and should normally only correct for small integration errors. It looks from your plots the integration and magnetometer have the opposite sign. So the fast initial response is measured by the gyros, but then the magnetometer measures the opposite and it takes 10 seconds before the filter has compensated this huge "integration error"


-Christophe 



On Fri, Oct 28, 2011 at 11:56 AM, Loic Drumettaz <address@hidden> wrote:
Hello,

I performed the calibration of aspirin IMU on a Lisa M board of the accelerometers and magnetometers, using the phyton program. For the gyrometers, I have tuned the neutral, but i have no possibility to calibrate the sensitivity. Then, I checked the phi, theta and psi angles using the AHRS telemetry message.
phi and theta are looking fine, but psi response is weird: see attached plot showing the psi angle measurement, moving the quadrirotor stepwize. The initial transient goes in the opposite direction of the movement, and the response time is 8 seconds! For AHRS, I'm using the "int_cmpl_euler" configuration.
I have investigated and plotted the FILTER message and it appears that this behaviour is due to the filtering process in the AHRS: INT32_measure_psi is OK. (except noise and offset...)
- Are there parameters available in Paparazzi for tuning the AHRS filter?
- In AHRS section of the airframe configuration file, what does PROPAGATE_FREQUENCY parameter stand for? I tried different values but didn't see any difference.
- Is this problem due to wrong measurement/ calibration of the magnetometer? 

Thanks in advance

Loïc


_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel



reply via email to

[Prev in Thread] Current Thread [Next in Thread]