C. 1995 
s for the 
Automatic 
& Space 
297-306. 
fusion for 
roceedings 
hiques et 
jour. PhD, 
“ontext in 
'S of the 
ge (MA), 
iting Map 
mage and 
JROPTO- 
FEATURE POSITIONING ACCURACY IN MOBILE MAPPING: 
RESULTS OBTAINED BY THE GPSVan™ 
John D. Bossler and Charles K. Toth 
Center for Mapping 
The Ohio State University 
1216 Kinnear Road, Columbus, OH 43212-1154 
e-mail: bossler@cfm.ohio-state.edu 
USA 
Commission IV, Working Group 1 
KEY WORDS: Accuracy, Feature Extraction, Real-time Technology, Mobile Mapping, GPS, GIS 
ABSTRACT 
Technology developed over the past five years at the Center for Mapping at The Ohio State University is 
being utilized for surveying in a variety of applications. One of the most demanding tasks required of this 
technology in the past year was to determine the coordinates of features to submeter accuracy along the 
Burlington Northern Railroad (BNR) right-of-way. Features include adjacent tracks, switches, and other 
items of interest to BNR. 
This paper discusses results obtained in this survey using the Global Positioning System (GPS) combined 
with dead reckoning (DR) (vector) observations, integrated with an imaging system. The image coordinates 
of features along the track are transformed to coordinates on the ground (object space). The "internal" 
accuracy of the photogrammetry is very good and the dominant error of the positions of features near the van 
is contributed by the GPS/DR system. To obtain the final system accuracy we compared the 
photogrammetrically derived coordinates of the features with ground truth acquired by using static GPS 
observations. 
acquiring spatially referenced digital data at normal 
highway speed from a vehicle is powerful. This feat 
is possible by integrating GPS, DR, and imaging 
The Center for Mapping at The Ohio State data to make a system that is as accurate as 
University (OSU) pioneered Mobile Mapping necessary for almost any application. New and 
technology, realizing that GISs require current, high — exciting applications for MMS are rapidly emerging 
quality spatial data for enhanced decision-making. ^ (Novak and Bossler, 1995). To discuss each one is 
The focus of this paper is to prove the assertion that — impossible. However, two key applications that are 
the GPSVan™ technology can obtain extremely briefly described are: E-911 and Facilities 
high quality data in an effective manner. Management. 
1. INTRODUCTION 
Mobile Mapping Systems (MMS) can be defined as E-911 
moving platforms upon which multiple There is a need to locate houses, driveways, phone 
sensor/measurement systems have been integrated to ~~ booths, fire hydrants, and other objects that will 
provide near-continuous positioning of both the ^ help emergency personnel (and their vehicles) as 
platform's path in space and other simultaneously — they come to the aid of callers. There are several 
collected geo-spatial data. As a result of GPS competing data acquisition systems, but a survey 
technology, MMSs have grown in use and with a hand-held GPS unit seems to be the most 
popularity over the past five years." The concept of — formidable competition to an MMS. However, the 
hand-held GPS unit cannot compete with the 
GPSVan™ technology, since one only needs to 
drive the van past the property to acquire 
  
! A recent two-day symposium in Columbus, Ohio sponsored by 
the Center for Mapping at The OSU drew 188 attendees. 
139 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996