Hi Balata,

surely, once, I finish my thesis, I will send a copy of it to you.

Thank you - don't forget it!

and you gave me a good indication of how precisely estimate.

I was surprised, how good the algorithm estimates the pyramid-surface and an uncertainty of about 2 mm seems to be realistic.

But I couldn't determine the precision of the apex. A copy of the second part of the program(computation of the apex) I attached below.

I don't study the complete code but it seems, that you estimate each plane in a single step and combine these solutions to derive the apex. The "problem" is the imperfection of the measurement. Therefore, it is possible that no intersection point is given by the 3 (or more) planes. Thus, the consideration was to formulate the apex as restriction and forcing an intersection. So, the plane parameters maybe deviate from a single solution.

This procedure is not new or developed by me. A similar problem was solved by Hermann Bähr: Präzise Vermessung des Phasenreferenzpunktes von Corner-Reflektoren pp. 13-24 (link to PDF).

Is [JAG3D] is also refer to your self?

I'm the developer, yes. I don't have a publication which focus the software itself because it is a leisure project. I introduce JAG3D at the last tech12-continuing education seminar:

Lösler, M. (2012). JAG3D – Ein kostenfreies Programm zur Netzausgleichung und
Deformationsanalyse. Neitzel, F. (Hrsg.): tech12 - Aktuelle Trends der Messdatenauswertung
in Kataster- und Ingenieurvermessung, 19./20. April 2012, Berlin, Deutschland.

But you can also write in a similar way to the pyramid:

Lösler, M. (2012). JAG3D – A free Program for Network Adjustment and Deformation Analysis.
http://derletztekick.com/software/netzausgleichung (last access: 2012-12-02).

Can I ask if possible to send to me the program of JAG3D for fitting cone and sphere instead of software in order to be able to use through matlab.

The source code is free. You can find it e.g. at the SVN on sf.net.

Together with Dr. Matrin Nitschke, we publish some Matlab/Octave code to fit a plane or ellipse but I never create a m-File for cone or sphere.

kind regards
Micha

--
applied-geodesy.org - OpenSource Least-Squares Adjustment Software for Geodetic Sciences