Procedures for the First and Second Order Design of Vertical Control Networks

Schriftenreihe
des Instituts für Geodäsie

Heft 7/1982


 
Procedures for the First and Second Order Design of Vertical Control Networks

Paul CROSS / Ali FAGIR
London, England

 

In: BORRE, Kai / WELSCH, Walter M. (Eds.) [1982]:
International Federation of Surveyors - FIG -
Proceedings Survey Control Networks
Meeting of Study Group 5B, 7th-9th July, 1982, Aalborg University Centre, Denmark

Schriftenreihe Wissenschaftlicher Studiengang Vermessungswesen, Hochschule der Bundeswehr München, Heft 7, Neubiberg, S. 159-179.
 



Abstract

Optimal network design problems can be posed in many ways but for the special case of level networks the following is most common.

GIVEN (a) the approximate position of the required bench marks,
  (b) a list of possible level routes,
  (c) the precision expected of the field parties and
  (d) the intended use of the network.
FIND   The set of level routes whose observation will yield a network suitable for the intenged purpose with the minimum of cost (i.e. minimum total length of levelling).

The above problem is of course, an example, of the now standard "first order design". To date no direct solution to this problem has been found but various iterative strategies using the second order design procedure (i.e. solution for observation weights and subsequent discared of observations with small weights) have been employed.

This paper critically reviews second order design procedures that have been applied to the design of vertical control networks and gives details of two sets of experiments with a new method which has a number of advantages including not needing the explicit fromation of a criterion matrix.
 



zurück zur Übersicht des Heftes 7
 


 

 

DOWNLOAD


 

"7-Cross-Fagir-159-179.pdf"

(311 KB)