additional constraint for the ambiguity
resolution. Due to these constraints the number
of potential solutions which have to be tested is
reduced by a significant amount and hence the
on-board processing requirements is minimal.
Further, the resolution of the ambiguities is
much more reliable as for the positioning case.
However, for the airborne, kinematic attitude
determination the baseline can not enter the
attitude computations as error-free. Due to
structural deformations of the aircraft body and
wings during the flight, caused by the
aerodynamic behaviour, the baseline length can
vary significantly (see also COHEN/PARKINSON
[1992]),
It has been mentioned earlier, that the highest
measurement accuracies are required for GPS
attitude determination. Therefore, error effects
like multipath, antenna phase center variations,
wing flex and receiver noise have to be kept as
small as possible. A more detailed review about
the error effects and the data-processing
techniques for real-time, kinematic attitude
determination with GPS can be found e.g. in
HARTL/WEHR [1986]. Due to its simple handling
its robustness and the minimal data processing
requirements GPS attitude determination is
highly suitable for real-time mapping
applications. The potential of this sensor has
been analyzed under conventional
photogrammetric conditions, yielding attitude
accuracies in the range of 0.1 - 0.2 degrees.
Figure 3 shows a comparison of GPS attitudes
and the attitudes computed from a conventional
photogrammetric block adjustment. A more
detailed description of this testflight can be
found in SCHADE ET AL. [1993].
1
[gon] A heading
067 € pliich
9 rol
02+ &
02+
0.8 +
1 : | | : |
375 395 415 435 455 475
[Image nr.]
[Figure 3 Differences between Photogrammetric
and GPS Attitudes
4. SUMMARY AND CONCLUSIONS
The usage of GPS for the real-time
determination of exterior orientation parameters
in the context of airborne mapping applications
has been discussed. From experimental, airborne
kinematic GPS data an accuracy potential for
real-time positioning of 0.1 - 5 m has been
achieved, depending on wether the correct
carrier phase cycle ambiguities can be
determined or not. Attitude parameters with
multi-antennae GPS receivers can be
determined in real-time with an accuracy of 0.1-
0.3 degrees. Although, the attitude accuracies
are not sufficient for the direct stereo-evaluation
of overlapping images, they can be used as
approximate values for image matching
techniques. On the contrary, the accuracy of the
attitude parameters may be sufficient for the
direct exterior orientation of non-imaging
sensors, monoplotting purposes or the direct
exterior orientation of airborne-remote sensing
sensors. In conclusion, it has become obvious
that with the advent of the Global Positioning
System the real-time exterior orientation of
airborne sensor systems can be achieved with
sufficient accuracy for most mapping
applications.
5. BIBLIOGRAPHY
BLACKWELL E.G. [1986]: Overview of
Differential GPS Methods, Global Positioning
System, Papers published in Navigation, Vol.
III, The Institute of Navigation, Washington,
1986
COHEN C.E, PARKINSON B.W. [1992]: Aircraft
Applications of GPS Based Attitude
Determination, Proceedings of the ION GPS-92,
Fifth International Technical Meeting of the
Satellite Division of the Institute of Navigation,
pp. 775-782, Albuquerque, USA 1992
FREI E., BEUTLER G. [1990]: Rapid Static
Positioning Based on the Fast Ambiguity
Resolution Approach "FARA": Theory and
Results, manuscripta geodaetica, Vol. 15,
Springer Verlag, pp. 325-356, 1990
188
HARTL
Transfor
Control,
I Sympo
ESA SP
1986
HATCH
Resolutic
107 Kin
and Rei
Canada,
KALAFU!
[1986]
Recomm
Global P
Navigati
Navigati
LACHAPI
FALKENI
Code T
Surveys,
Geodetic
Columbi
SCHADE
Errors i
Ambigui
Archive:
Sensing,
SCHADE
An Acct
Attitud
NAVSTA
Zeitschri
Navigati
90-95,1¢