ters in 
image- 
ing of 
van, 
St was 
points 
of the 
. The 
ors for 
nputed 
ove the 
les are 
ate that 
tem is 
of the 
Cy we 
1e wall 
system 
S were 
natation 
amera 
‚Asa 
;parate 
mmon 
ioning 
ws the 
mined 
  
[cm] : [em] i [em] : [em] 
Sz Sy Oz 
  
  
  
stereo-pair object distance 
[m] 
6 23.0 
7 17.0 
  
4.0 
14.5 
16 . 1959 
  
  
  
  
79 
  
Table 3: Comparison of point positioning in a local coordinate system. sy, Sy, Sz are the 
mean differences between the positions computed from stereo-pairs 6 and 7. ©, is the 
theoretical accuracy limit for a given object distance (compare table 2). 
As the global van locations and the orientation angles 
are available after post-processing of GPS and inertial 
observations, the local object coordinates can be 
transformed into a global coordinate system. Again, the 
point coordinates obtained by two different stereo-pairs 
were compared. The results are shown in table 4. One can 
see that the global positioning accuracy is somewhat lower 
than the local accuracy, however, better than the expected 
GPS position-accuracy. We believe, that the 
transformation parameters between camera coordinates and 
global coordinates could still be improved. 
  
  
  
  
  
units Sx Sy Sz 
degrees 0.00000356 0.00000023 28.8 [cm] 
centimeters 39.6 2.6 28.8 [cm] 
  
  
  
Table 4 : Comparison of point positioning in a global coordinate system. The global 
coordinates were obtained in degrees (longitude, latitude) and meters (height). For 
easier comparison they were converted to centimeters. 
6. CONCLUSIONS 
The calibration of all sensors and their application to 
transform local, spatial coordinates in to a global system 
are essential tools of the GPS-Van. Without knowing the 
calibrated parameters with a very high accuracy the stereo- 
vision system would not be useful. Therefore, these 
functions are forming the core of our post-processing 
system. The other important component of positioning 
with a digital stereo-vision system however, is automatic 
image analysis. A variety of functions have been 
implemented on our post-processing workstation to extract 
features such as road-edges and traffic signs, and to follow 
lines or to match points. These techniques are subject of 
another paper of this conference (He, et. al.; 1992). 
The initial GPS-Van resulted in a number of follow-on 
projects and developments. We are currently working on 
the integration of GPS and a digital mapping camera in an 
aircraft (MAPCAM). Here we apply conventional 
photogrammetric triangulation and positioning to digital 
imagery. As a next step we will integrate a digital camera- 
pair in an airplane together with three GPS receivers. This 
will be used to map power-lines and gas pipelines; it is 
called Utility Mapping System (UMS). Finally, we 
designed a portable Digital Stereo-Positioning System 
(DSPS), which consists of two cameras and three GPS 
receivers. It can be set up by the user on a tripod to capture 
an image-pair. As both position and attitude of the DSPS 
are known at any time (both from GPS), every object in the 
field of view of the cameras is immediately available in a 
world coordinate system. From the positive reaction and 
interest by private companies and government agencies, 
and the successful demonstration of the GPS-Van we 
conclude that the application of real-time mapping systems 
is almost unlimited, and that they will revolutionize mobile 
mapping. 
7. REFERENCES 
Bossler J., Goad C., Johnson P., Novak K., 1991. “GPS and 
GIS Map the Nation’s Highways.” Geolnfo 
Systems Magazine, March issue, pp. 26-37. 
Brown D.C., 1976. "The Bundle Adjustment - Progress and 
Prospects." Invited paper XIII th Congress of ISP, 
commission III, Helsinki. 
Goad C., 1991. “The Ohio State University Highway 
Mapping System: The Positioning Component.” 
Proceedings of the Institute of Navigation 
Conference, Williamsburg, VA, pp. 117-120. 
He G., Novak, K., 1992."Automatic Analysis of Highway 
Features from Digital Stereo-Images." 
International Archives of Photogrammetry and 
Remote Sensing, Vol. , Commission III. 
Novak K., 1991. “The Ohio State University Highway 
Mapping System: The Stereo Vision System 
Component.” Proceedings of the Institute of 
Navigation Conference, Williamsburg, VÀ, pp. 
121-124. 
8. ACKNOWLEDGEMENT 
The authors wish to thank all transportation agencies 
that committed money to the development of the GPS-Van. 
We gratefully acknowledge the exciting research 
environment and great support of the GPS-Van team at the 
Center for Mapping of the Ohio State Univesity.