GSI Forum: Bugs, Fixes, Releases

Home » PANDA » PandaRoot » Bugs, Fixes, Releases » KalmanFit Error

Show: Today's Messages :: Polls :: Message Navigator

KalmanFit Error [message #5376]

Fri, 09 November 2007 09:59

Tobias Stockmanns
Messages: 489
Registered: May 2007

first-grade participant

From: *ikp.kfa-juelich.de

Dear KalmaFitters,

when I try to fit MVD data with the Kalman task, I get the error, that (most of) my matrices are zero, which I do not understand.

Maybe one of you can help me Smile

Thanks

Tobias

MvdKalmanTask::Exec
starting track0
-I- MvdRecoHit::MvdRecoHit(MvdHit*) called.
o: 0.287864 2.69613 5.975
u: 0.707107 -0.707107 0
v: 0.707107 0.707107 0
-I- MvdRecoHit::MvdRecoHit(MvdHit*) called.
o: 2.11 1.67845 8.735
u: 0 1 0
v: 1 0 0
-I- MvdRecoHit::MvdRecoHit(MvdHit*) called.
o: 2.20361 1.60633 14.3662
u: 0.00398854 -0.999595 0.0281843
v: 0.999992 0.00399013 0
-I- MvdRecoHit::MvdRecoHit(MvdHit*) called.
o: 4.63742 3.15743 18.475
u: -0.729407 0.68408 0
v: 0.68408 0.729407 0
4 hits in track 0
starting fit
Error in <TDecompLU::DecomposeLUCrout>: matrix is singular
Error in <TDecompLU::InvertLU>: matrix is singular, 0 diag elements < tolerance of 2.2204e-16
FitterException thrown with whatString:
cannot invert covsum in Kalman Gain - det=0
in line: 265 in file: /home/stockman/fairroot/cbmsoft_old/pandaroot/genfit/Kalman.cxx
FitterException Info Output
===========================
---------------------------
Matrices Label String:
cov, HitCov, covsum1 and covsum
---------------------------
Matrices:

5x5 matrix is as follows

| 0 | 1 | 2 | 3 | 4 |
----------------------------------------------------------------------
0 | 0 0 0 0 0
1 | 0 0 0 0 0
2 | 0 0 0 0 0
3 | 0 0 0 0 0
4 | 0 0 0 0 0

3x3 matrix is as follows

| 0 | 1 | 2 |
--------------------------------------------
0 | 0.002887 0 0
1 | 0 0.002887 0
2 | 0 0 0

5x3 matrix is as follows

| 0 | 1 | 2 |
--------------------------------------------
0 | 0 0 0
1 | 0 0 0
2 | 0 0 0
3 | 0 0 0
4 | 0 0 0

3x3 matrix is as follows

| 0 | 1 | 2 |
--------------------------------------------
0 | 0.002887 0 0
1 | 0 0.002887 0
2 | 0 0 0

===========================

Report message to a moderator

Re: KalmanFit Error [message #5381 is a reply to message #5376]

Sat, 10 November 2007 10:41

Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich

first-grade participant

From: *cern.ch

Hi Tobias

To me your error looks as if you did not setup your initial covariance matrix properly. Make sure you initialize at least the digonal elements with some (large) values.

Cheers, Sebastian.

Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592

Report message to a moderator

Previous Topic:	segfault in dch macros
Next Topic:	Error in charmtask

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sun Nov 03 20:53:43 CET 2024

Total time taken to generate the page: 0.00897 seconds