DIMU with very short full calibration results

Info & Discussions about good ole AQ6, taken to the next level.

Re: DIMU with very short full calibration results

Postby joebar.rc » Sat Feb 22, 2014 7:05 am

afernan wrote:@Menno:
thanks. The AltHold is AWESOM! I´m testing indoor (lot of room turbulences) and I can´t believe the precission! This is a great improvement.

Also the SIMPLE leveling is very useful. Unfortunally is not kept after switch off.
Angel


I know :), alt hold on baro only is almost perfect.
Regarding the leveling bias not storing, well it should but in case you're quite pleased with the result simply dont switch off, connect to the GCS and do a permanent store to memory (all param widget).

I have some changes made to the d_imu.c that allows me to save and recall (based on a param value) to the dimu eeprom.
I did a test to see if the calibrated values stored in the eeprom can be used on an other board.
Simply placed the dimu on that board, use the eeprom values and it almost did fly the same. (temp compensation is different obviously, but not very promiment)

Menno
joebar.rc
 
Posts: 1084
Joined: Fri Jun 22, 2012 8:59 am
Location: The netherlands

Re: DIMU with very short full calibration results

Postby afernan » Wed Feb 26, 2014 3:08 pm

Calibration with DIMU
I´m trying to follow your instructions, but still some problems. I´m using command line.

- If I use the <cal,sim3, & cal_version=2.0> supplied by Menno from this post it fails in step3 and on. I´m using this command with only 1 static file:
cal --mag -p DIMU.params static1.LOG : din0.LOG

It goes well if I use old exe files.
Is it still valid using old cal, sim3 files?
Can we mix results from different dynamic files in different steps?

Attached the best results found. Are they useful?

STEP 1: MAE = 0.0006
STEP 2: MAE = 0.02, AVG = 9.80
STEP 3: MAE = 0.55, AVG = 1.93
STEp 4a: MAE = 0.18
STEp 4b: MAE = 0.11
STEp 5: MAE = 0.13
STEp 6: MAE = 1.82
Attachments
DIMU_AFD.zip
(1.61 KiB) Downloaded 76 times
afernan
 
Posts: 1214
Joined: Sat Jul 07, 2012 5:24 pm
Location: Madrid, Spain

Re: DIMU with very short full calibration results

Postby sandmen » Wed Feb 26, 2014 3:39 pm

You are using the binary's from this thread?
sandmen
 
Posts: 997
Joined: Fri Jun 22, 2012 7:25 am

Re: DIMU with very short full calibration results

Postby afernan » Thu Feb 27, 2014 9:54 am

Yes.
afernan
 
Posts: 1214
Joined: Sat Jul 07, 2012 5:24 pm
Location: Madrid, Spain

Re: DIMU with very short full calibration results

Postby hogster » Wed Apr 16, 2014 11:48 pm

Hi all :)

Can I just confirm that to get past step 3 without it getting stuck in a loop (which is what mine is doing now), I need to download the files Menno posted here:

viewtopic.php?f=42&t=2813&start=10#p17081

They appear to be .exe files though – how do I use those on my Mac?

These are the values which caused the mag calculations to start looping ...


MAE: 0.0237059
AVG: 1.98528
lowMAE = 0.023706, n[0] = 0.000033
#define IMU_MAG_BIAS_X +3.580864193e-02
#define IMU_MAG_BIAS_Y -2.215527153e-03
#define IMU_MAG_BIAS_Z -5.915706104e-02
#define IMU_MAG_BIAS1_X -1.035584623e-04
#define IMU_MAG_BIAS1_Y -3.915437575e-05
#define IMU_MAG_BIAS1_Z +1.738318160e-04
#define IMU_MAG_BIAS2_X +9.171626288e-05
#define IMU_MAG_BIAS2_Y -2.285913228e-05
#define IMU_MAG_BIAS2_Z -9.210397261e-05
#define IMU_MAG_BIAS3_X +2.635885416e-06
#define IMU_MAG_BIAS3_Y -9.131014038e-08
#define IMU_MAG_BIAS3_Z -1.944534783e-06
#define IMU_MAG_SCAL_X +2.801508199e-02
#define IMU_MAG_SCAL_Y +8.191101717e-03
#define IMU_MAG_SCAL_Z +6.505807639e-02
#define IMU_MAG_SCAL1_X -6.495802860e-05
#define IMU_MAG_SCAL1_Y -9.653673602e-05
#define IMU_MAG_SCAL1_Z -1.778187829e-04
#define IMU_MAG_SCAL2_X +7.356665601e-05
#define IMU_MAG_SCAL2_Y +6.756844251e-05
#define IMU_MAG_SCAL2_Z +1.035732574e-04
#define IMU_MAG_SCAL3_X +2.044946997e-06
#define IMU_MAG_SCAL3_Y +7.655600456e-07
#define IMU_MAG_SCAL3_Z +2.108898718e-06
#define IMU_MAG_ALGN_XY -5.692055017e-02
#define IMU_MAG_ALGN_XZ +9.781556637e-02
#define IMU_MAG_ALGN_YX -5.907953562e-02
#define IMU_MAG_ALGN_YZ -5.275796960e-02
#define IMU_MAG_ALGN_ZX +3.078151250e-01
#define IMU_MAG_ALGN_ZY -1.579771072e-01
#define IMU_MAG_INCL -6.616999817e+01


Should I be ok to move onto the next step with those?

Many thanks,

David :)
hogster
 
Posts: 239
Joined: Tue Feb 11, 2014 8:59 am

Re: DIMU with very short full calibration results

Postby teramax » Thu Apr 17, 2014 12:36 am

i am with you Menno, baro alt hold works awesome indoors, in fakt gps outside is killing the baro performance a little :)

anyway, your quad performs very good with the big props on, would you share your pid´s for comparison with us?
somehow everything can fly
http://www.youtube.com/user/ter4m4x
teramax
 
Posts: 500
Joined: Sun Sep 02, 2012 11:19 pm
Location: Lower Austria

Re: DIMU with very short full calibration results

Postby hogster » Thu Apr 17, 2014 6:18 pm

Hmm I moved onto step 4a, but the MAE value jumped up and down from the start. After 100 loops it's currently reading:

1.443
1.436
1.440
1.395
1.421
1.445
1.434
1.418
1.423
1.452
1.446

It doesn't seem to be settling down towards any particular value, nor is it continuously decreasing either ...

I've tried restarting the calculation but it does much the same thing each time.

There seem to be some developments on the whole calibration side of things at the end of this thread:

viewtopic.php?f=42&t=3075&start=60

so I might just give up on the calibration for now ...

Any input is appreciated as always :)

Many thanks, David
hogster
 
Posts: 239
Joined: Tue Feb 11, 2014 8:59 am

Re: DIMU with very short full calibration results

Postby hogster » Sun Apr 20, 2014 8:49 am

Happy Easter everyone! :)

Does anyone have any recommendations for my next steps with the calibration, re: the above posts? Should I wait until the new firmware is developed?

Many thanks! David
hogster
 
Posts: 239
Joined: Tue Feb 11, 2014 8:59 am

Re: DIMU with very short full calibration results

Postby teramax » Sun Apr 20, 2014 10:55 am

Happy Eastern also to Every one.
I can tell ya, first load Jussis defaults, do the flat freeze and warmup log, do the dance.
For calculation ( are the new Sim in the qcg package already in cludet?) Make cal 1 and 2 like before, cal 3 whatch out we're it stops with nan, and repeat. But stop shortly before numbers came near the last run endet with nan. All other steps do like you made in the past with the IMU.
Watch out your numbers and horizon and don't forget the easter bunny came across this morning :D
somehow everything can fly
http://www.youtube.com/user/ter4m4x
teramax
 
Posts: 500
Joined: Sun Sep 02, 2012 11:19 pm
Location: Lower Austria

Re: DIMU with very short full calibration results

Postby hogster » Sun Apr 20, 2014 1:17 pm

Thanks for the reply! Sounds like you've had too many chocolate Easter bunnies though ;)

Do my results above look ok? The values in step 4 didn't seem to be tending towards a stable value so I stopped it after 200 or so cycles – the result keeps floating up and down. Is this normal or does it point to a problem with my data?

Many thanks, David :)
hogster
 
Posts: 239
Joined: Tue Feb 11, 2014 8:59 am

PreviousNext

Return to AQ6 +DIMU +CAN

Who is online

Users browsing this forum: No registered users and 2 guests

cron