Fw: Differential Orbit Correction

From: Greg Roberts via Seesat-l <seesat-l_at_satobs.org>
Date: Tue, 24 Jun 2014 11:18:54 +0200
Morning all
With my continual cloud and rain at the moment chances of optical observing are
rare so Im playing around with some computer software.

 Some of the old timers may recall the Satellite Orbits group that operated from
 the Royal Aircraft Establishment at Farnborough  and issued a regular
 rhoneographed publication called SATELLITE OBSERVING NOTES and from time to 
 had articles contributed by observers in the OBSERVERS FORUM. One such article
 in April 1979 was by David Hopkins who was the first amateur to observe and
 maintain orbits on classified earth satellites.

 His article contained a program listing of the program DOC. The original
 computer listing appears to have been retyped onto a wax stencil by someone and
 in the process introduced several errors etc with sometimes the script running
 out of page, and sometimes rather illegible.

 I tried tackling this program in 1988 and gave up with the comment on the
 original script "Impossible to debug - too full of errors"!  In a moment of
 weakness a few days ago I decided to tackle this problem again and have now
 corrected etc about 90% of the original script but have now hit a serious bug
 and, for the moment, it looks like Im doomed.

 The program basically forms 7 pairs  for RA and Dec for each of the 7 elements
 of the orbit which may be changed - time is kept as a constant. This data is
 used to form 7 simultaneous equations in the seven unknowns needed to match the
 residuals between computed and observed positions and the equations are solved
 using a matrix inversion.

 The original program was written in BBC Basic (?) and occupied about 2K and
 mostly used single letter variables - sometimes using a particular variable to
 mean something else from what it was originally. It was also written
 specifically for Davids station at Bournemouth so there were a lot of constants
 in the program . I have managed to identify all constants used and rewrite the
 program for multi-station use.

 As I said earlier Ive got the program to succesfuly run through about 90% of 
 rewritten code but have reached a point where it appears a matrix ends up by
 being zero and is used a short while later to be a divisor and of course
 anything divided by the value zero causes the program to crash.

 So my query now is -

 (1) did anyone else try and use /rewrite this program

 (2) were any program corrections issued in Satellite Observing Notes subsequent
 to the listing that appeared in SATELLITE OBSERVING NOTES 725 of 1979 ?   I
 unfortunately no longer have these issues.

 I last heard from David Hopkins about ten years ago at which point he had given
 up satellite tracking and taken up terrestial bird watching instead.

 Any help/comments would be greately appreciated and yes, I know Im reinventing
 the wheel but I enjoy programming and hopefully prevents me from going senile 
soon :-))


Seesat-l mailing list
Received on Tue Jun 24 2014 - 04:19:28 UTC

This archive was generated by hypermail 2.2.0 : Tue Jun 24 2014 - 09:19:29 UTC