Christian Heipke
2 TEST OBJECTIVES AND EXPECTED RESULTS
The European Organisation for Experimental Photogrammetric Research (OEEPE) has embarked on a test investigating
integrated sensor orientation using GPS and IMU in comparison and in combination with aerial triangulation. The test
has been set up as a multi-site test. The Institute for Photogrammetry and Engineering Surveys (IPI), University of
Hanover acts as pilot centre for the test. Data acquisition and pre-processing including the organisation of test flights
and the necessary field work was carried out by the Department of Mapping Sciences (IKF), Agricultural University of
Norway in As.
The main focus of the test is on the obtainable accuracy for large scale topographic mapping using photogrammetric
film cameras. The accuracy of the results will be assessed by investigating the exterior orientation elements of an image
block and in particular with the help of independent check points on the ground in the following scenarios:
= aerial triangulation,
= GPS/IMU observation for the projection centres only,
.* combination of aerial triangulation with GPS/IMU.
The test is expected to demonstrate to which extent integrated sensor orientation using GPS and IMU with and without
aerial triangulation is an accurate and efficient method for the determination of the exterior orientation parameters for
large scale topographic mapping. A comparative analysis will show in which way the mentioned potential error sources
and an integrated bundle block adjustment, with or without using a minimum of ground control points have an influence
upon the accuracy of the derived orientation parameters and ground control coordinates. Furthermore, the potential and
problems of integrated sensor orientation will be assessed based on the comments of the participants. A detailed inves-
tigation into the transformation of the raw GPS and IMU measurements into flight trajectories and attitude values (roll,
pitch, and yaw as a function of time), however, is out of the scope of this test. Rather, the flight trajectories and the
attitude values as computed by the GPS/IMU systems, are considered to be the input of the test, and are to be processed
together with the image data.
3 TEST DATA SET
3.1 Selection criteria and general description
The test will be carried out based on especially acquired imagery and GPS/IMU data. In order to enable a fair and
meaningful test between the two competing technologies the following selection criteria for the data acquisition were
set forward:
= geometrically stable photogrammetric block,
= modern photogrammetric film camera,
= dual frequency GPS receivers using differential carrier phase measurements with a data rate of 0.5 sec, pref-
erably identical receivers for the plane and reference station,
= a short base line between plane and reference station,
= high quality off-the-shelf navigation grade IMU as typically used in precise airborne attitude determination,
= different image scales suitable for large scale topographic mapping,
= a well-controlled test field with a large number of ground control points.
Given these criteria and a few practical constraints a test field in Fredrikstad, Norway, was selected. Two companies
producing suitable GPS/IMU equipment agreed to participate in the test, namely Applanix of Toronto, Canada, using
their system POS/AV 510-DG (Hutton J., Lithopoulos E. 1998; Applanix 2000), and IGI mbH of Hilchenbach, Ger-
many, with the system AEROcontrol II (IGI mbH 1999).
3.2 Test field Fredrikstad
The test field Fredrikstad lies in the south of Norway near the capital Oslo. It is maintained by IKF. The test field has
already been used in a prior OEEPE test on GPS photogrammetry (Andersen, Ackermann 2000), its size is approxi-
mately 5 x 6 km. 51 well distributed signalised ground control points with UTM/EUREF 89 coordinates and ellipsoidal
heights known to better than 0.01 m are available. The ground control point targets have a size of 40 x 40 cm”.
374 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000.
