required to transmit all the phase observations within the
measurement update rate.
For the tests which have been carried out in the scope of
this paper no radio link was used. The coordinates which
have been derived from the real-time code solutions of
the airborne receiver were compared with the results
from the aerial triangulation. The differences between
these solutions are shown in figures 3a and b for the two
receiver types.
30
20 4-4
10.
Difference [m]
Image Nr.
[-]
Figure 3a Differences (Xy,z) between Aerial
Triangulation and Real-Time Computed Positions (Leica
9212 Aero)
30.
20.
10. of
0 à
Difference [m]
Image Nr.
[-]
Figure 3b Differences (x,y,z) between Aerial
Triangulation and Real-Time Computed Positions (Leica
SR 399)
For the Leica 9212-Aero a root mean square value of
16.24 m has been achieved over all 3 coordinate
components (x,y,z). As expected the performance of the
SR 399 is slightly better (r.m.s. = 14.25), because the
receiver is capable of reducing the effect of the
ionosphere using the observations on both frequencies.
Also, it is expected, that the signal to noise ratio of the
SR 399 is superior to the one of the 9212-Aero receiver,
as special techniques allow P-code aided tracking on
both frequencies. Similar accuracies could have been
achieved in further tests under varying conditions. Apart
from the above mentioned features the dual frequency
receiver does not have apparent advantages over the
732
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B3. Vienna 1996
cheaper single frequency navigation receiver. When
comparing the achieved results with the demands for
photoflight navigation and automatic camera release, it
can be seen, that using the tested hardware the given
accuracy requirements can be met even for large scale
applications.
6. ACCURACY OF POST-PROCESSED DGPS
CAMERA COORDINATES FOR REDUCTION OF
GROUND CONTROL
Besides navigation and camera control, GPS is used
more frequently for the determination of the camera
perspective centers. The economic benefits, due to a
significant reduction of ground control points have
convinced photogrammetrists to use post-processed
GPS observations in the block adjustment. Although, the
majority of the GPS error sources can be eliminated by
differential positioning, the code observations of modern
receivers still can not provide the required centimeter
accuracy for large scale applications. To achieve this
accuracy it is necessary to use the GPS carrier phase
observations. However, the problems with the phase
observations is, that it is necessary to determine the
correct set of cycle ambiguities in order to exploit the
inherent accuracy of a few millimeters. Several methods
have been proposed to fix the cycle ambiguities in an
airborne environment. These methods will be briefly
reviewed here:
1. As long as no loss of lock or cycle slips occur, the
ambiguities remain constant integer values. In
principle, they can be estimated and fixed in a static
initialization at the beginning of a continuous
trajectory, but due to banking angles in flight turns
and the highly kinematic environment losses of
phase lock and cycle slips are frequent in
photogrammetric applications.
2. Ambiguity resolution on the fly (AROF), tries to
resolve the carrier phase ambiguities from the GPS
data alone. Sophisticated statistical tests are used
to distinguish between the correct and incorrect
ambiguity sets.
3. The third method is the combined adjustment of
GPS and photogrammetric image data (CBA). The
ambiguity resolution is done in a two step
procedure. In a first step the ambiguities are fixed
only roughly in a GPS post-processing step and the
final determination of the camera perspective center
coordinates is done in the block adjustment using
the image coordinate measurements and the post-
processed GPS positions (FrieR [1990],
Ackermann/Schade [1993]).
It is obvious, that the first method has lost importance,
due to the doubtful reliability and the unfavorable
economic aspects as flat turns increase the flying time
considerably.
Today, photogrammetrists are usually choosing between
the methods 2 and 3. Under geodetic conditions
ambiguity resolution on the fly has already reached an
operational status (see e.g. Frei/Beutler [1990], Hatch
[1990]). The statistical tests which are used in AROF
obs
con
Unti
Suc:
pho
The
envi
The
noc
amb
a ble
up s
The
imac
aerie
pers
GPS
bloc!