The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
1209 
The exposure of the images is done mechanically whereby an 
electrical impulse triggers a mechanical device which in turn 
triggers the camera. Approximate values for coordinates (X, Y, 
Z-position) of the perspective centre and the flight direction k 
are obtained via a GPS-logger (GlobalSat DG 100) which 
records a GPS NMEA-string every second. 
Figure 2: UAV “SUSI” in the air 
After the flight the GPS-data and the recorded trigger impulses 
are synchronised. During the flight the GPS-data is down linked 
to ground and serves for the navigation of the UAV on a laptop. 
On the laptop the GPS-data is displayed on top of a 
georeferenced map or an aerial map to support the navigation 
and the triggering of the images once the UAV is within the 
survey area. The most important technical parameters of the 
two micro-UAVs are summarised in Table 1. 
UAV Carolo P330 
UAV „SUSI“ 
Type of aircraft 
Model plane 
Paraglider UAV 
Weight 
5 kg (max. payload 
0.4 kg) 
5 kg (max. payload 
5.0 kg) 
Speed 
16 m/s - 30 m/s 
0 m/s - 8 m/s 
Range 
+++ 
++ 
Endurance 
max. 60 min 
max. 140 min 
Weather and Wind 
dependency 
++ 
+++ 
Sensor platform 
Fixed, camera inside 
model plane 
Gimbal-mounted 
platform 
GPS transfer / 
recording 
Downlink and 
onboard storage 
Downlink and no 
onboard storage 
Synchronisation 
GPS/camera 
Not available 
Not available 
Sensor 
Canon PowerShot 
S60 
Sony DSC R1 
Sensor size (calc.) 
7.176 * 5.319 mm 
21.5 * 14.4 mm 
Resolution (pixel) 
2,592 * 1,944 
3,888 * 2,592 
Pixel size (calc.) 
2.7 pm 
5.5 pm 
Type of chip 
CCD 
CMOS 
Exposure interval 
fixed, every 5 s 
manually 
Exposure delay 
not applied (~ 0,15 s) 
not applied 
Navigation 
autonomously (Way 
Points) 
Manually (display on 
PC) 
Table 1: Comparison of the two Micro-UAVs “Carolo P330” 
and “SUSI” 
3.3 Interior Orientation 
The determination of the interior orientation is necessary for 
both Micro-UAV’s which use off-the-shelf digital cameras. The 
calibration was done with the software “Australis” , Version 
6.0 from Photometrix (www.photometrix.com.au). Internally 
the software is based upon a free network optimisation. The test 
field calibration was done with a flat test field. The test field, 
which consists of 35 retro targets has an extent of approx. 3.5 * 
4.0 m. Images were taken from 5 different positions. A total of 
14 converging images were taken for a calibration. The 
software computes the full range of the interior orientation 
parameters. Because for the later processing the 
photogrammetry software LPS was be used, a calibration with a 
limited set of parameters (focal length, image center, radial 
distortion) was conducted. A graphic display of the radial 
distortion of both cameras (Figure 3) shows that the Sony DSC 
R1 is nearly free of radial distortion, while the Canon 
PowerShot reveals a very strong radial distortion of up to 100 
pixels at the image comers. 
Radial distance r' [mm] 
Figure 3: Radial distortion of Canon PowerShot S60 and Sony 
DSCR1 
4. RESULTS 
4.1 Test flights 
A total of three test flights were conducted to obtain practical 
information of the photogrammetric performance of the two 
Micro-UAVs. Two test flights were carried out with the System 
Carolo P 330 on 4 th May 2007. Beside the pthotogrammetric 
analysis the data should be used for a vitality analysis of field 
trials at the University of Applied Science in Soest. The 
weather conditions for the flights were quite suitable apart from 
a wind of ca. 3 Bft. Due to the small size of the first test site 
(200 x 450 m) Merklingsen 1 (M 1 in table 2) the first flight 
was conducted manually. The second test site Merklingsen 2 
(M 2) with an area of 1000 x 750 m was flown autonomously. 
The test flight Wahlsdorf (W 1) (300 x 500 m) with the system