acc table idea...

Static, Dynamic, Log File Analysis, Troubleshooting

acc table idea...

Postby gionag » Tue Nov 27, 2012 2:24 pm

bill take in my mind some elucubration about the dynamic calibration.

as stated before, you can split the process of dyna calibration in two parts, one for the gyo/acc and one for mag.
the gyo/acc can be accomplished by only turning the board itself rather than the whole multi.

so i start thinking about making some sort of gimbal for the board, driven by two 360 servo on tilt and roll axes.

maybe an arduino with the right library for the servo where you can program the movements as the calibso dance and then they will be reproduced with all the smoothness and precision possible.

i just started coding something and today i wish to finish the frame of the gimbal.

just one thing.... the 360 continous movements is useful for the gyro / acc ? i think the contiunous turning movement is to be intended for mag purpose. am i right ?

Brainstorming time :mrgreen:

after some testing i'll certanly post my results.
gionag
 
Posts: 85
Joined: Mon Oct 22, 2012 3:12 pm

Re: acc table idea...

Postby bn999 » Tue Nov 27, 2012 4:55 pm

The reason for the continuous Z axis rotation during the dance is so that you get a fair chance of exposing all of the octants of the imaginary cube to the Earth's magnetic vector in your area. The axis of the MAG vector is, of course, offset from the gravity vector. If you are not interested in calibrating the MAGs (as in you are doing the first of two separate dances) the Z axis rotation is not needed.
bn999
 
Posts: 1559
Joined: Thu Jun 21, 2012 11:40 pm

Re: acc table idea...

Postby gionag » Tue Nov 27, 2012 8:22 pm

exactly what i'm thinking...

what do you think about the gimbal for calibrating acc/gyro ?
gionag
 
Posts: 85
Joined: Mon Oct 22, 2012 3:12 pm

Re: acc table idea...

Postby bn999 » Tue Nov 27, 2012 8:28 pm

If it is done properly, then it could make things easier for sure. Servos tend to be jerky and possibly too fast.

You will need 3 axes to get all of the required GYO data.
bn999
 
Posts: 1559
Joined: Thu Jun 21, 2012 11:40 pm

Re: acc table idea...

Postby gionag » Wed Nov 28, 2012 12:18 am

jerky but still better than hands :D

maybe i'll try some stepper motor....
gionag
 
Posts: 85
Joined: Mon Oct 22, 2012 3:12 pm

Re: acc table idea...

Postby nikivan » Wed Nov 28, 2012 1:19 am

Why go into so much trouble? The calibration only takes few minutes, and you'd spend considerably more time trying to build a solution that does not even include all steps (mags).

If you have experience with stepper motors, it would be very helpful to research their use in high-precision camera gimbal. DJI incorporates them in the Zenmuse unit, which beats anything available today for the price.

Just my $0.02.
nikivan
 
Posts: 78
Joined: Wed Jul 04, 2012 12:32 pm

Re: acc table idea...

Postby Max » Wed Nov 28, 2012 1:40 am

Sounds like a perfect job for some little brushless motors and a custom controller... something like this (Peter was just telling me about it the other day) http://fpv.no-comp.net/arduino/fpv-vide ... ess-motors

I think the only real reason(s) to bother would be if it made the gyro calibrations more accurate (or possible, in the case of a big vehicle), or if you had to calibrate a lot of boards... ;)

-Max
Max
 
Posts: 2814
Joined: Mon Aug 13, 2012 9:45 pm
Location: Near Ithaca, NY, USA


Return to Calibration

Who is online

Users browsing this forum: No registered users and 16 guests

cron