Summary
Aerial triangulation is of very great importance for the densening of
the geodetic network, particularly in connection with topographic mapping.
The accuracy of the results of aerial triangulation is evidently of basic
importance for the practical applications. Up to now the accuracy has
mainly been determined by empirical methods only. The results are conse-
quently dependent wpon the circumstances and conditions which were present
during the empirical tests. In this paper an attempt is made to find general
analytical expressions for the accuracy which can be expected from some
different methods for aerial triangulations in strips and under well defined
assumptions and approximations. Strip triangulation is of particular im-
portance for road and highway planning. Block triangulation is usually
founded upon strip methods.
Introduction
In this paper there will be treated first the theory of errors of the
stereo-radial method (HALLERT 1957 a and 1957 b) followed by a similar
treatment of the independent model triangulation method (EKELUND
1950 and 1951—52). Finally ordinary aerial strip triangulation (v.
GRUBER 1935, BACHMANN 1946) will be investigated.
The following assumptions are made. First it is assumed that the
most important systematie errors of the fundamental operations are
carefully determined and corrected for. Consequently, the residual
errors are assumed to be mainly of accidental character. The statistical
expression for these errors is the standard error of unit weight of the
fundamental measurements (image coordinates and parallaxes) according
to the method of the least squares.
The aerial photography is assumed to be approximately vertical and
performed under normal uniform conditions with comparatively small
differences of ground elevation (less than about 15 7/ of the flying
altitude). Control points are available only in the beginning and in the
end of the strip. No redundant control is assumed to be available.
According to fig. 1 the strip consists of the individual photographs
1, 0, 1, 2,.... 9 n, l, n, n + 1.
