GSI Forum: Tracking » GEANE backward propagation

Home » PANDA » PandaRoot » Tracking » GEANE backward propagation

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

GEANE backward propagation [message #6963]

Wed, 18 June 2008 10:07

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

first-grade participant

From: *cern.ch

Hi Lia!

I use GeaneTrackRep to do some backward propagations to extroplate Tracks back to the z-Axis / Target. A lot (>50%) of these propagations simply fail and I do not know why. I cannot see any clear pattern which tracks are affected.

Are you aware of such problems?

How can we find out why the propagation fails?

Cheers! Sebastian.

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

Report message to a moderator

Re: GEANE backward propagation [message #6967 is a reply to message #6963]

Wed, 18 June 2008 11:08

Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino

first-grade participant

From: *PV.INFN.IT

Hi Sebastian,
are you using the GeaneTrackRep::extrapolate(const DetPlane& pl, TMatrixT<double>& statePred, TMatrixT<double>& covPred, TMatrixT<double>& jacobian)) function? Because in that function I see that when backprop is true you call _geane->setBackProp() which sets fPropOption="BLE" but in that case GEANE would make the back-propagation to a specific tracklength ("L" option), while what we want here is to back-propagate to a plane: so in this case it should be fPropOption="BPE".

Hope this helps,
Lia.

Report message to a moderator

Re: GEANE backward propagation [message #6969 is a reply to message #6967]

Wed, 18 June 2008 11:18

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

first-grade participant

From: *cern.ch

Hi Lia!

That is already an good hint! How can I set the correct option?

Yet for my problem at hand I am mostly using GeaneTrackRep::getPocaOnLine where I call _geane->BackTrackToVirtualPlaneAtPCA(2); And this causes the problems. Is there a similar thing here?

Cheers!
Sebastian.

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

Report message to a moderator

Re: GEANE backward propagation [message #6975 is a reply to message #6969]

Wed, 18 June 2008 13:15

Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino

first-grade participant

From: *PV.INFN.IT

Quote:

How can I set the correct option?

I think the CbmGeanePro function setBackProp() has been created by you, so I don' t exactly now where you use it (only here or somewhere else). My suggestion is that if you use it only here (where the "BPE" option is required) you can change directly that function, while if you need the setBackProp() function to set "BLE" somewhere else, you could add a new function to CbmGeanePro, something like setBackPropToPlane {fPropOption = "BPE";}.

Concerning the _geane->BackTrackToVirtualPlaneAtPCA(2) I used it in stt local Kalman and it worked fine for me... it sets the propagation in two steps: first it finds the PCA with the CbmGeanePro::FindPca(...) function, it creates the virtual plane here and then it makes the actual propagation (the PropagateFromPlane is set internally, it is not necessary to set it before).

Does it fail while finding the PCA (in FindPca, when it calls ERTRAK) or later when doing the propagation (in the usual Propagate(PDG) function)?

Ciao,
Lia.

Report message to a moderator

Re: GEANE backward propagation [message #6976 is a reply to message #6975]

Wed, 18 June 2008 14:58

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

first-grade participant

From: *cern.ch

Hi!

It fails in the usual Propagate Function. That means the function returns false.

Cheers!
Sebastian.

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

Report message to a moderator

Re: GEANE backward propagation [message #6977 is a reply to message #6976]

Wed, 18 June 2008 15:09

Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino

first-grade participant

From: *PV.INFN.IT

Hi Sebastian,
I just uploaded to svn a fixed CbmGeanePro: in the old one there was a too restrictive check for orthogonality of the virtual plane axes. Please try updating your geane directory and check if the failure happens again so often.
Ciao,
Lia.

Report message to a moderator

Re: GEANE backward propagation [message #6978 is a reply to message #6977]

Wed, 18 June 2008 15:44

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

first-grade participant

From: *cern.ch

Hi Lia! You are a darling!
That was it! Now it works quite well!!!

Thanks!
Sebastian.

PS: I will check in an updated version of the setPropDir function in cbmGeanePro later.

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

Report message to a moderator

Previous Topic:	Update GEANE to use q/p instead 1/p
Next Topic:	cov[0][0] = 0 in Kalman.cxx

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Thu Nov 28 01:02:04 CET 2024

Total time taken to generate the page: 0.00717 seconds