Design of a Mobile Mapping System 
for GIS Data Collection 
Guangping He 
Lambda Tech International, Inc. 
W239 N1812 Rockwood Dr. Suite 100 
Waukesha, WI-53188 
USA 
Data acquisition is the most expensive part of establishing 
a Geographic Information System (GIS). A mobile 
mapping system, the GPSVision, which integrates the 
Global Positioning System (GPS), an Inertial Navigation 
System (INS) and stereo cameras, has been developed to 
quickly and accurately collect the digital data of civil 
infrastructures. While the mobile mapping system drives at 
a normal speed, the GPS and INS record the position and 
attitude information of the system, and the stereo vision 
system takes image pairs at regular interval. After 
GPS/INS data is processed, the position and orientation of 
every image pair is determined in the global coordinate 
system. Any features which are visible on an image pair 
can be located in a global coordinate system. 
The GPS provides accurate position data. Because of low 
data rates and a requirement of viewing at least four 
satellites, the use of GPS alone is limited. In contrast, the 
INS provides high rate position and attitude information, 
but its sensor errors tends to grow with the time. By 
integrating GPS and INS, the accurate GPS position is 
used to update the INS, and the INS then produces high 
rate, accurate position and attitude data, even when the 
GPS signals are lost. 
From an application perspective, a mobile mapping system 
can be used to collect stereo digital images along 
highways, state roads and residential streets, while 
traveling at posted speed limits. These digital images are 
accessed by feature extraction software to locate the 
positions of visible physical facilities, such as manholes, 
curb lines, traffic signs, pedestals and building locations. 
The positions and attributes of these visible features are 
stored in a simple format which is transportable to 
standard GIS systems. Once the processed data is loaded 
into the target GIS, the data is displayed in map format 
and manipulated utilizing database query functions. A 
typical client could use these data to accurately position 
traffic signs, develop base maps or view image data as one 
drives down the road. 
This paper will discuss the basic issues of a mobile 
mapping system. A brief description of the system 
configuration is given, followed by the integration of 
GPS/INS using a kalman filter, and system calibration for 
determining the offsets between the different sensors. How 
the position and attributes of a feature is collected and 
transferred into a GIS are addressed. Automatic 
154 
information extraction from image pair is also discussed. 
Finally, the data sets collected by a mobile mapping system 
are presented to demonstrate the efficiency and 
applications of the mobile mapping system. 
1. Introduction 
Generation of the civil infrastructure is an immense task. It 
requires efficiently collecting vast quantities of data. New 
technologies offer opportunities to greatly improve our 
effectiveness. One of the most innovative technologies for 
collecting civil infrastructure data is the GPSVision -- a 
Mobile Mapping System developed by Lambda Tech 
International, Inc. The GPSVision quickly acquires highly 
detailed, multimedia data about highways, railroads and 
other transportation networks. It uses the Navstar Global 
Positioning System(GPS), integrated with an Inertial 
Navigation System(INS) along with a state of art color 
CCD cameras. The result is that any feature (e.g. a 
manhole) which is “seen” by two cameras can be precisely 
located in a global coordinate system. 
A GPS receiver determines the global location of the 
system. Depending on the type of the receiver and the 
processing software used, positioning accuracy can range 
from meters to centimeter. Because obstructions such 
bridges, trees, tunnels or high rise buildings can interrupt 
satellite signals, GPS alone can not meet the requirement 
for a mobile mapping system. An INS consists of the 
accelerometer and gyroscopes. It is a self contained 
system. It measures the velocity and direction changes 
very accurately for short periods, but its error grows 
rapidly with time. An integration solution of GPS and INS 
can greatly improve the system performance. GPS is used 
to update the INS system and the INS outputs the 
accurate position, velocity and attitude of the system 
between the GPS updates. 
The stereo vision system consists of two high resolution 
color CCD cameras. It captures stereo image pairs while 
the system drives at posted speeds. The images are taken 
according to a distance interval to provide a full coverage 
of the road way and its surroundings. By applying 
photogrammetric triangulation technology, any point that 
appears in both images can be located into a global 
coordinate system. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B2. Vienna 1996 
In or 
calibi 
lens « 
offse 
syste 
const 
may 
syste: 
is th 
contr 
of th 
copla 
the p 
used. 
suite 
Data 
direct 
infori 
comp 
The 
multi 
data 
consi 
a lang 
The | 
GPSY 
ProX 
navig 
GPS" 
of G 
GPS 
requis 
achie 
impor 
platfc 
vehic| 
and tl 
  
A PC 
displa