The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B5. Beijing 2008 
less than about 70 km from the nearest reference station (Hutton, 
J. et al, 2007). 
2. TEST RESULTS 
This section presents results from a number of studies into the 
new capabilities of POSPac MMS. 
2.1 BLOM Case Study 
Using data collected from a LIDAR calibration flight conducted 
in April 2007, BLOM Geomatics AS of Norway performed a 
study on POSPac MMS to investigate the accuracy using 
existing reference stations, and the potential for flying sharp 
turns. 
2.1.1 Accuracy Using Existing Reference Stations: In 
order to test the accuracy of the Applanix SmartBase module, 
an existing set of 27 reference stations located throughout 
Norway, Denmark and Sweden was used to create four separate 
Applanix SmartBase networks, with baselines 60, 110, 200 and 
300 km apart respectively (Figures 4 and 5). 
The coordinates of each reference station were transformed 
from ETRS89 to ITREF 2000 epoch 2007.30 in order to place 
them in the same frame of reference and epoch as the GPS 
observables, and the SmartBase quality check adjustment was 
run. The adjustment detected that one of the reference stations 
was using an incorrect antenna model. This was subsequently 
fixed and the Smartbase solution was computed for each 
network, which were then used to generate 4 different SBET 
solutions using the Applanix IN-Fusion processing in POSPac 
MMS. 
The SBET solutions for each network were then differenced 
with a reference or “truth” solution generated using the loosely 
coupled POSPac V4.4 processing and a network of 7 reference 
stations that included a dedicated base station directly in the 
project area. The accuracy of the POSPac V4.4 reference 
solution was estimated to be 4 cm RMS horizontal and 5-7 cm 
RMS vertical for position, and 20 to 30 arcsec for the 
orientation. The results revealed that once the solution had 
converged, the position differences with the reference for the 60, 
110, and 200 km networks were all well below 10 cm, and the 
orientation differences were well below 30 arcsec (Figures 6 
and 7). Only for the 300 km network did the differences start to 
grow as the software had trouble resolving the correct 
ambiguities (Figure 8). 
Figure 6. Position Differences, 200 km Network 
Figure 7. Orientation Differences, 200 km Network 
831