The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Voi. XXXVII. Part B5. Beijing 2008 
In 1979, an aeroplane, a UAV platform with fixed wings was 
firstly used for photogrammetry by Wester-Ebbinghaus (1980). 
But it was found that it is hard for the platform to approach a 
target. Hence, helicopters, UAV platforms with rotating wings, 
have been utilized for taking the close images of a target 
(Wester-Ebbinghaus, 1980). 
Zischinsky, et. al. (2000) utilized a helicopter to acquire 82 
ground pictures and 38 aerial pictures acquired and created 3D 
building modelings. Nagai (2004) generated a DSM from the 
data acquired by a laser Scanner and a CCD camera mounted 
on a Subaru helicopter whose maximum payload weight was 
100kg. Jang (2004) used a mini UAV-helicopter to take the 
images of ancient historic scenes. Wang (2004) developed mini 
aeroplane to extract 3D building model from 2D GIS data and 
one image. Everaerts (2004) designed a UAV-system which 
was called pegasus. This UAV-system was equipped with solar 
cells to fly longer time. 
Most systems developed from these previous studies did not 
have real-time (or rapid) mapping capability, which can 
produce spatial information such as DSM and orthoimages of 
the target area under an emergent situation. For the realization 
of this capability, the raw sensory data acquired by a UAV 
based multi-sensor system should be transmitted to a ground 
station and automatically processed. These functions were not 
fully integrated into the most of previous systems. 
In the meantime, the individual technologies required for the 
realization of the UAV based rapid mapping system have been 
rapidly advanced in different fields. For examples, many robust 
automatic processing algorithms have been developed for the 
generation of spatial information from sensory data such as 
aerial images and airborne LIDAR data in the photogrammetry, 
remote sensing, and computer vision fields. In addition, based 
on the advances in the fields of aviation and navigation, small 
helicopters with a self-flying capability at a close range have 
emerged. In summary, many individual technologies in the 
fields of geo-spatial information, aerospace engineering and 
electronics have been enough matured to be used for the 
implementation of real-time rapid mapping system. 
To realize the real-time rapid mapping for the practical uses in 
many emergent situations, we plan to develop a light, flexible 
and low-priced mapping system using a small unmanned 
helicopter. As distinct from the existing UAV-systems based 
expensive hardware, our system will be based on highly 
sophisticated software instead of using a high-priced platform 
and sensors. As being in an initial stage of developing this 
system, we will present an overview of this project, including 
the preliminary design, work breakdown structures, expected 
product and others. 
2. PROJECT OVERVIEW 
2.1 Goals 
The goal of this project is to develop a light and flexible system 
to perform rapid mapping for emergency responses. This 
system consists of two main parts, aerial part and ground part. 
The aerial part is composed of a UAV platform, sensors, and 
sensor supporting modules. The mounted sensors are GPS, IMU, 
digital camera and laser scanner. The sensor supporting 
modules undertake to integrate the sensors, transmit the sensory 
data to the ground, and stabilize sensors’ attitudes. The ground 
part is composed of three sub-systems with appropriate 
software, which are a control/receiving/archiving sub-system, a 
data geo-referencing sub-system, and a spatial information 
extraction sub-system. The whole system is illustrated in Figure 
2. 
Figure 2. Introduction to real-time aerial mapping system 
The sensory data acquired from the aerial system are images, 
returned laser information, position and attitude of the platform 
tagged with GPS time. These data will be transmitted to the 
ground system through the sensor supporting modules. Then, 
images and laser points autonomously will be geo-referenced 
by the geo-referencing sub-system of the ground system. 
Finally, from the geo-referenced sensory data, the spatial 
information extraction sub-system autonomously will generate 
DSM/DEMs and orthoimages of the target areas where disaster 
or accident occurs. 
2.2 Configuration 
This project is one of a series of projects supported through 
Korean Land Spatialization Group (KLSG) funded by Korean 
government. The overall budget is about 6 million US dollars 
and the period is about four years. 
This project is divided into 5 sub-projects, as listed in Figure 3. 
We are concentrating on the first sub-project, design of real 
time aerial monitoring system for the first period (2007.11.30 ~ 
2008.8.30). Five sub-projects are grouped by research items 
into three sectors, aerial sector, ground sector, system 
verification as shown in Figure 4. 
iiiticA« [ t ш fi Ш i Ligi Ш ш ‘Чи i h i'i hi f it hi rit t 1« 't 
Prototype Development 
, iti E f ulMcst* §if i t [-Г, ggltf* gill f i 
m 
for Real-time Transmission of Sensor Data 
evelopment of Image Acquisition Stabilization System 
\ Generation through Multi-sensor fusion 
Technology and its Applications 
Figure 3. List of sub-projects