cal --acc fails with "nan" in AQGC v1.0.2 - fixed in 1.0.4

Ground Stations and utilities for Windows, Linux, Android, & others

cal --acc fails with "nan" in AQGC v1.0.2 - fixed in 1.0.4

Postby Max » Mon Sep 03, 2012 1:59 pm

Running parameter generation with latest downloadable version of QGC with AQ widget 1.0.2 has a problem with Step 2. Using older version, 1.0.1, works fine with the same log files. I've attached the output copy/pasted from the UI. It starts out fine, and then all the numbers show up as "nan" (I'm assuming that's "not a number") and it terminates quickly.

I redid the Calibso while waiting for the older version to download and tried the new (smoother) log file with 1.0.2, but same result. Step 1 worked fine in 1.0.2.

Is the newer version of cal.exe better in some way? Should I wait until 1.0.2 works, or is it OK to generate all the params using 1.0.1?

If there's anything I can do to help troubleshoot this, just let me know.

Thanks,
-Max

PS. Forum doesn't allow uploading .txt files? :o
Attachments
QGC102 step2 cal nan output.zip
(1.02 KiB) Downloaded 159 times
Last edited by Max on Wed Oct 10, 2012 1:20 pm, edited 1 time in total.
Max
 
Posts: 2814
Joined: Mon Aug 13, 2012 9:45 pm
Location: Near Ithaca, NY, USA

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby sandmen » Wed Sep 05, 2012 7:28 am

Do you get it working?
sandmen
 
Posts: 997
Joined: Fri Jun 22, 2012 7:25 am

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby Max » Wed Sep 05, 2012 7:43 am

Nope, 2 machines, reboots, different dynamic logs... nothing works with cal --acc included in the 1.0.2 download. I ran step 2 in 1.0.1 and using that output saved to params file in 1.0.2 for the other steps.

I'm getting this during the sim3 --acc run and I'm wondering why the ACC_ALGN_* values are all zeros and if it could be related to my cal --acc troubles?

Code: Select all
0.039304 0.009048 0.062865 0.000000 0.000000  = 0.111217
Loops:  851

#define IMU_ACC_BIAS_X      -1.637100051102   //  0.000000000009 +0.000000282013
#define IMU_ACC_BIAS_Y      -1.637312848052   //  0.000000000009 +0.000000778271
#define IMU_ACC_BIAS_Z      -1.617229956499   //  0.000000000008 +0.000000508953
#define IMU_ACC_SCAL_X      +0.017946886149   //  0.000000000010 -0.000000020658
#define IMU_ACC_SCAL_Y      +0.018219973087   //  0.000000000010 -0.000000019867
#define IMU_ACC_SCAL_Z      +0.019999174195   //  0.000000000010 -0.000000029550
#define IMU_ACC_ALGN_XY      +0.000000000000   //  0.000001000000 +0.000000000000
#define IMU_ACC_ALGN_XZ      -0.000000000000   //  0.000001000000 -0.000000000000
#define IMU_ACC_ALGN_YX      +0.000000000000   //  0.000001000000 +0.000000000000
#define IMU_ACC_ALGN_YZ      +0.000000000000   //  0.000001000000 +0.000000000000
#define IMU_ACC_ALGN_ZX      -0.000000000000   //  0.000001000000 +0.000000000000
#define IMU_ACC_ALGN_ZY      +0.000000000000   //  0.000001000000 +0.000000000000


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

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby sandmen » Wed Sep 05, 2012 7:46 am

can you send me some small sample of your log files ?
(one static and one dynamic)
Thanks
sandmen
 
Posts: 997
Joined: Fri Jun 22, 2012 7:25 am

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby Max » Wed Sep 05, 2012 7:56 am

I'm uploading them now and will post a link here (my upload speeds aren't so good). Meanwhile, I think I figured out why my ALGN values are zero...

Step 2 (using cal from v.1.0.1) had generated this (in part):

Code: Select all
#define IMU_ACC_ALN_XY +7.254134188e-003
#define IMU_ACC_ALN_XZ -7.712227429e-002
#define IMU_ACC_ALN_YX +5.729517457e-002
#define IMU_ACC_ALN_YZ +1.980468352e-001
#define IMU_ACC_ALN_ZX -6.315427578e-002
#define IMU_ACC_ALN_ZY -9.673650163e-002


Shouldn't "ALN" be "ALGN"?

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

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby Max » Wed Sep 05, 2012 8:26 am

Here you go:

http://max.wdg.us/AQ/Max_Dynamic_02.zip (15MB)
http://max.wdg.us/AQ/Max_Static_02.zip (125MB)

Changing "ALN" to "ALGN" in the params file fixed my sim3 --acc output. I guess that was a bug with the older cal.exe?

Not sure if you've seen my other thread about sim3 --acc crashing unless I use a variance of 1e-8... don't know if this is all related somehow.

Thanks for looking into it!
-Max
Max
 
Posts: 2814
Joined: Mon Aug 13, 2012 9:45 pm
Location: Near Ithaca, NY, USA

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby sandmen » Wed Sep 05, 2012 8:55 am

#define IMU_ACC_ALN_XY +7.254134188e-003
#define IMU_ACC_ALN_XZ -7.712227429e-002
#define IMU_ACC_ALN_YX +5.729517457e-002
#define IMU_ACC_ALN_YZ +1.980468352e-001
#define IMU_ACC_ALN_ZX -6.315427578e-002
#define IMU_ACC_ALN_ZY -9.673650163e-002

yes, that's the problem.
I think also for the sim3.
Please start with the latest qgroundcontrol, the calculation again.
sandmen
 
Posts: 997
Joined: Fri Jun 22, 2012 7:25 am

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby Max » Wed Sep 05, 2012 10:48 pm

Hi Peter,

Just saw the addition of the last 2 lines of your last post. I think there's some confusion (at least for me).

The problem you quoted (ALN should be ALGN) is with the output of cal.exe from version 1.0.1. I can't get 1.0.2 cal --acc to run, which is why I used the older version for that step. So the problem was just for the sim3 --acc step and I don't how it would solve the cal.exe "nan" problem.

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

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby Max » Fri Sep 07, 2012 1:23 am

Peter,

I've confirmed that, as luck would have it, the Static_02 log file I uploaded is the cause of these "nan" errors, and in fact all my calculation problems so far. I hope it is useful in tracking down the errors.

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

Re: cal --acc fails with "nan" in AQGC v1.0.2

Postby epyonxero » Sat Sep 08, 2012 7:13 pm

Im having a similar problem. When I load all 3 of my static log files the output of Step 2 is "nan". But if I remove any one log file and just run with two I get an actual output.
epyonxero
 
Posts: 193
Joined: Thu Aug 30, 2012 4:09 pm

Next

Return to AQ Software

Who is online

Users browsing this forum: No registered users and 6 guests

cron