Re: Workshop on Automatic Beamline Alignment

[Grant Bunker <bunker@biocat1.phys.iit.edu>]

We (BioCAT) intend to move in this direction, but aren't very far along in
implementing it yet. 

grant 

On Wed, 16 Dec 1998, Tim Mooney wrote:

> Anybody interested in this?  I'm pretty sure John Quintana has
> something like this running; I don't know about anyone else.
> By the way, Olof's abstract is still online at
> http://www.aps.anl.gov/xfd/bcda/nobugs/svensson.html
> 
> 
> ------------- Begin Forwarded Message -------------
> 
> >From svensson@esrf.fr Wed Dec 16 06:34 CST 1998
> To: 'Tim Mooney' <MOONEY@aps.anl.gov>
> Cc: 'Andy Thompson' <thompson@esrf.fr>
> Subject: Workshop on Automatic Beamline Alignment
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> X-Priority: 3
> X-MSMail-Priority: Normal
> X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0
> 
> Dear Tim,
> 
> I guess you remember me, we met during NOBUGS II last year where I gave a
> presentation on Automatic Beamline Alignment (ABA). I'm together with the
> ESRF research directors organizing a workshop on ABA. The workshop will take
> place here at the ESRF on March 12th and 13th 1999. The idea of the workshop
> is to discuss ABA in general and in particular ongoing projects in different
> facilities, and to initialize collaborations in order not to reinvent the
> wheel.
> 
> I'm wondering if you, or someone else, is working on ABA at the APS and
> would be interested to give a presentation in the workshop. If so, please
> confirm before the end of this week. I need the names in order to send out
> an official invitation in the beginning of next year. The ESRF can pay the
> travel, hotel and meals for two persons from the APS.
> 
> Regards,
> 
> Olof Svensson
> email: svensson@esrf.fr
> 
> 
> 
> ------------- End Forwarded Message -------------
> 
> 
> 
> My own plans in this area are to develop an EPICS record to implement
> several minimization algorithms (e.g., random, simplex, gradient) using
> N positioner links, 2 detector-trigger links, M detector links (sound
> familiar?) and a calc expression.  To the user, this record would look
> pretty much like the scan record, but instead of scanning over a grid
> within preset limits, the objective would be to search within those limits
> for the combination of positioner values that minimizes the result of the
> calc expression.  At the moment the project is welded securely to the back
> burner.
> 
> Tim Mooney
>