forum.thequadshot.com
http://thequadshot.com/forum/

Expresso: drifting in mode 1
http://thequadshot.com/forum/viewtopic.php?f=5&t=834
Page 2 of 2

Author:  david.erzen [ Sun May 11, 2014 7:09 pm ]
Post subject:  Re: Expresso: drifting in mode 1

Hey,

regarding the drift: if I want to "calibrate" the sensors manually in order to cancel the drift, which parameters should I change and which parameters will tell me that the change was good?

I'm thinking of

-parameters to change: ACCEL_X_NEUTRAL, ACCEL_Y_NEUTRAL, ACCEL_Z_NEUTRAL, BODY_TO_IMU_PHI, BODY_TO_IMU_THETA, BODY_TO_IMU_PSI

-parameter to check: STAB_ATTITUDE (?)

Any suggestions?
Thnx,

David

Author:  david.erzen [ Sun May 18, 2014 3:29 pm ]
Post subject:  Re: Expresso: drifting in mode 1

Hi

I was playing with the parameters BODY_TO_IMU_THETA, BODY_TO_IMU_PHI in order to correct the drifting in hover mode.
I managed to decrease drifting rate by setting the BODY_TO_IMU_PHI to -6 degrees and BODY_TO_IMU_THETA to 2 degrees.
( I was performing indoor tests of the QS.)

Although the drift rate has decreased I think this is not the correct way to do it.

So my questions are:
- do the numbers indicate that the imu is not perfectly aligned with the body axis when hovering?
- what is the correct procedure to decrease drifting.

Regards,

David

Author:  david.erzen [ Mon Jun 02, 2014 11:55 am ]
Post subject:  Re: Expresso: drifting in mode 1

anyone?

Author:  psinha [ Wed Nov 12, 2014 5:25 am ]
Post subject:  Re: Expresso: drifting in mode 1

Hey David,
Do you mean the estimate was drifting or that the vehicle itself was drifting?
The procedure you describe indicates that the lisa is not sitting vertical in the hover mode. In this case, the thing to do is to either set it right in hardware by re-seating the lisa, or, an easier way is to do what you did and fix the rotations in software.
The physical drift will never be 0 without feedback control on the position. If you are interested in trying more precise positioning on the Quadshot, please shift to the master branch and use the kalman filter based estimator. This is a lot more stable. If you need help with this, we can setup a call.

Page 2 of 2 All times are UTC
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/