| operational
ogrammetric
e, real-time
required for
curate post-
GPS based
> operational
provided by
5 based
al
ngulation
0.01-1m
]hly dynamic |
st-Processing
yes
10-500 km
min. 1 Hz
uired in large
scale
pplications
"hase only,
rier smoothed
code
> major
pecific needs
remainder of
irdware and
nportance for
PECTS
It for specific
ise static (or
amic or low
applications,
ich has been
applications.
to choose a
J software is
ly dynamic
requirements
ot standard
today's GPS
ceivers and
cision results
ather difficult,
d the quoted
nderstood or
ay's receiver
Dierendonck
Today, normally two types of receivers are used for
photogrammetric applications:
e dual frequency geodetic receivers
e single frequency navigation receivers
As it can be seen from Table 2 each of the two receiver
types have certain advantages and disadvantages. The
major difference in the receivers and also in the price, is
the inherent positioning accuracy, and their ruggedness
in an airborne environment. In principle the geodetic
receivers can provide more accurate positioning results,
both in real-time and post-processing, as the ionospheric
error effects can be eliminated using a linear
combination of L1 and L2 observations. Also, the dual
frequency observations are the key to resolving
ambiguities on the fly. However, it is necessary to
discuss the necessity of this feature in connection with
photogrammetric applications more specifically (see
Chapter 6).
is used. Phase observations are availbale on both
frequencies (Jackson et al. [1995]). On the other hand
the 9212-Aero is a receiver mainly designed for
navigation in a rugged dynamic environment. It is a
continuous 12 channel single frequency receiver, giving
L1 carrier phase and C/A-Code observations.
The data which was used in this analysis is from two
testflights which have been carried out over the
photogrammetric test fields in:
e BUCHS, close to the Leica Factory in
Switzerland
e OHIO, a photogrammetric testfield in the
vicinity of Columbus, Ohio, USA
The most important parameters for these testflights are
summarized in Table 3.
Test Ohio Test Buchs
Image Scale 1:8000 1:4000
# of Lines 3 3
# of Images 14 20
Forward Lap 80 % 60 %
Side Lap 30 % 30 %
Camera Leica RC 30 Leica RC 30
# of control 42 57
points
GPS receiver Leica SR 399 Leica 9212-Aero
Anti-Spoofing On On
Distance from « 20 km « 80 km
Reference
Station
Geodetic Dual Single Frequency
Frequency Navigation
Receiver Receiver
Observation [L1: carrier, C/A- L1: carrier, C/A-
Types Code, P-Code Code
L2: carrier, P-Code
Accuracy low noise, by low to medium
narrow correlator noise with narrow
or P-Code tracking | correlator tracking
lonosphere lonospheric free can only be
linear combination modeled
Tracking usually very narrow medium
bandwidth
Measurm. usually 1Hz at least 1 Hz
Frequency
Ambiguity Possible under most likely not
Resolution optimal conditions | possible with GPS
On the Fly data alone
Channels typically 12 typically 6-12
continuous (,all in continuous
view")
Electromagn sensitive usually not very
etic Impact sensitive
Price range high low to medium
Table 2 Performance characteristics of GPS receivers
used in photogrammetry
4. DESCRIPTION OF THE TEST DATA
To assess the performance potential of modern GPS
receivers under photogrammetric conditions a series of
tests has been carried out with two receivers which are
used in photogrammetry today. The SR 399 is a full dual
frequency receiver, which has been mainly designed for
high precision geodetic applications. It provides C/A-
Code observations on the L1 frequency with a noise
reduction using the narrow correlator technique. P-Code
observations are available even under Selective
Availability as a proprietary, patented P-Code technique
731
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B3. Vienna 1996
Table 3 Blockparameters for Testflights Ohio and
Buchs
The coordinates of the camera perspective centers which
are taken as reference for the accuracy analysis are
derived from a conventional aerial triangulation. In the
1:4000 block Buchs the estimated co for the perspective
center coordinates is 2.3 cm and for the Ohio block the
Oo is 4.9 cm.
5. ACCURACY OF REAL TIME POSITIONING FOR
NAVIGATION AND CAMERA CONTROL
As it has been mentioned above, photo flight navigation
and automatic camera release requires real-time
positioning. Today, the real-time position computations
are usually based on code observations from the
airborne receiver alone. Normally, no radio links are
used to increase the positioning accuracy using real-time
differential GPS. However, the situation may change, as
wide area augmented GPS networks and additional
satellite systems are currently beeing built up (McLellan
et al. [1994], Till et al. [1994]), to provide GPS correction
signals with standardized communication protocols,
allowing for real-time differential code positioning. It will
take another few years until real-time differential phase
positioning becomes feasible for airborne applications.
The limiting factor is the high transmission rate, which is
