The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
1230 
mentioned processing chain are recapitulated. The paper 
concludes with an outlook on ongoing and future development 
steps. 
2. UNMANNED AERIAL VEHICLE SYSTEMS 
2.1 Overview 
A rapid development of unmanned aerial vehicle systems has 
taken place in the last few years. Today, a wide variety of 
different UAV systems exists on the market. The European 
Association of Unmanned Vehicle Systems (EUROUVS) has 
drawn up a classification of the different system platforms. In 
(Bento, M. 2008) the current EUROUVS classification is 
presented and a good state of the art overview is given. 
Category 
Max. Max. Endurance Data 
Take Off Flight Link 
Weight Altitude Range 
Micro 
Mini 
< 5kg 250m lh < 10km 
< 30kg 150-300m <2h < 10km 
Table 1: Classification of mini and micro UAV systems 
This paper is focused on mini and micro UAV systems. Table 1 
shortly recapitulates the technical specifications of these two 
UAV categories, again based on the current classification by 
UVS International. Most of the mini or micro UAV systems 
available today integrate a flight control system, which 
autonomously stabilises these platforms and also enables the 
remotely controlled navigation. Several systems additionally 
integrate an autopilot, which allows an autonomous flight based 
on predefined waypoints. These flight control systems are 
typically based on MEMS (Micro-Electro-Mechanical System) 
IMU systems, navigation-grade GPS receivers, barometers, and 
magnetic compasses. The different sensor observations are 
usually integrated to an optimal flight state using an EKE 
(Extended Kalman Filter), which is subsequently used in the 
flight controller. For mapping applications, it is also possible to 
use this flight control data to geo-register the captured payload 
sensor data like still images or video streams. However, as a 
result of the utilisation of low-weight and low-cost flight 
control sensors, the achievable geo-referencing accuracy is 
strongly limited. 
2.2 Microdrones md4-200 platform 
Figure 1: Quadcopter micro UAV (microdrones md4-200) 
system (left) with portable ground control station (right) 
For the prototype solution presented in this paper we use the 
micro UAV platform microdrones md4-200 which is illustrated 
in Figure 1. The following listing contains a short overview of 
the technical specifications of the platform and of the sensors 
used for capturing video streams and flight control data. 
Platform 
UAV category: 
max. take off weight: 
max. payload: 
endurance: 
Sensors 
GPS receiver: 
IMU: 
magnetic compass: 
barometer 
video camera (payload): non-metric / PAL output 
resolution: 640x480 pixels 
Flight attitude accuracy (After sensor data fusion) 
position: 2.5 m CEP 
altitude: 5 m SEP 
roll and pitch angle: 1-2° 
yaw angle (Heading): 3-5° 
Notable is the low attitude accuracy of the integrated flight data, 
especially the heading accuracy which is approximately three 
times lower than that in roll and pitch. This has a direct 
influence on the achievable video geo-referencing accuracy, 
especially for large image to object distances. Details of the 
implemented sensor data fusion approach are presented in 
(Meister et al., 2007). The systems platform includes an 
analogue data link between platform and ground control station 
for video- and flight control data transmission. The flight 
control state consisting of position, velocity and attitude of the 
platform together with a time stamp in UTC is transmitted with 
4-5Hz. Due to the restricted payload of only 300g, it is not 
possible to use a video camera with high quality optics or a 
genlock capability. 
3. VIRTUAL GLOBES 
3.1 State of the art 
Different web-based 3D geoinformation services based on 
virtual globes are available today. Google Earth and Microsoft 
Virtual Earth are only two prominent examples. Most of the 
available virtual globe technologies have the possibility to 
integrate large amounts of geospatial content, like terrain 
models, orthomosaics, 3D objects, points of interest or 
multimedia objects. Many solutions have an excellent ability to 
stream very large volumes of geodata. 
Among the shortcomings of virtual globe services are the often 
outdated geodata contents. However, for many possible 
application scenarios like real-time surveillance or decision 
support applications the availability or integration of up-to-date 
or even live imagery data is crucial. Furthermore, with most 
virtual globes, it is not known which underlying earth model is 
implemented. However, for an accurate geospatial content 
integration the knowledge of the used geodetic global reference 
system(s) is crucial. Overviews of available virtual globe 
technologies can be found in (Thalmann, 2007) or (Bleisch and 
Nebiker, 2006). 
micro (quadcopter) 
0.9 kg 
0.3 kg 
20 min 
u-blox (navigation grade - 
pseudorange processing) 
6DOF MEMS based 
three-axis sensor