coordinate reference system, is no serious problem at all.
It may be solved by just some ground control points. Also
drift or systematic errors are not too serious. They may be
modelled mathematically by simplified relations, the addi-
tional parameters of which are handled per flight line as
unknowns and solved by the combined block-adjustment, as we
have seen in the statoscope case. It is only the internal
precision thereafter which is of direct interest. It will be
very much higher than the usual figures given for the abso-
lute accuracy of navigation systems. It is said, for in-
stance, that GPS will give direct positional accuracy in the
order of 10 - 20 m. It is quite certain, however, that in
our concept the accuracy will be easily 5 or 10 times better.
Also, when phase measurements are taken in addition to
pseudo-range measurements, the internal precision of GPS is
expected to be in the order of 1 m and may reach eventually
} «dm, especially Hf dn addition a stationary receiver is: app-
lied. Also the angular attitude data are expected to reach
soon internal precision in the order of 107*to 10-5 ,i. e.
10 - 1 mgon or better.
It is not our topic here to discuss the actual performance
of navigation systems. At present investigations and empiri-
cal tests are being prepared to study in particular the error
properties of GPS data. We can safely assume that high pre-
cision relative position data from GPS will be available in
near future, which will not depend on the high precision
P-code.
3.3 Let us therefore assume that future navigation systems
will deliver positioning and/or attitude data for each camera
station which are precise enough to contribute to the analy-
tical orientation of photographs. In order to combine them
with block-adjustment they may be treated as "observations",
of course after some preliminary reduction and transformation
into the approximate block coordinate system. We then obtain
additional observation equations for each camera station j
of the following type (here the letter a is used for the
observed auxiliary orientation data):
au; + Vox = 0; &[aY agi a,x + )
ap; * Vo, = $; £5h (bg i wibby R os )
gli toute (15116 2104 (e ob In )
i + d = Xo + ( qi + d, x + )
oj ! Yo ^ 'oj ts Sog? ex + )
$203 + Voi z Loi tuni A + f X + )
