912 
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Voi. XXXVII. Part B5. Beijing 2008 
these areas was chosen considering the possibility to perform 
accurate RTK survey. 
«J5S * J-im m . - a * '»• 
Figure 6. Example of ground point used for the comparison 
with GPS/RTK survey identified on the images acquired by the 
Road-Scanner system 
2.5 Data processing 
GPS data of MMS are processed by kinematic differential GPS 
approach, using the software POSPAC LAND 5.0, developed 
by Applanix that provides also the combination with the IMU 
sensor using a Kalman filter,. The reference station was located 
on a known vertex in a barycentre position respect the 
underground track. ETRF89 coordinates of this vertex 
belonging to the National Reference Network were used for the 
adjustment of the survey. 
3. SOME RESULTS 
3.1 Productivity in the Mobile Scanning Technique 
From the productivity point of view the Mobile Scanning 
Technique (i.e. MMS with laser-scanner) has produced relevant 
results. In fact it is possible to survey thousands of points along 
a road, running at a speed closed to the service rate, in a tight 
time. The experiences made on non-urban roads have shown 
that MMS can survey 50-100 km. of road per day, in 
back/forward direction. 
Hence this approach can be considered a good and cheap 
alternative to the traditional GPS survey, in particular for 
surveys along the roads, and it is the ideal tool for producing as- 
built drawings. 
3.2 Coordinates internal comparison 
As already mentioned here above, in order to check the 
obtained accuracy and the repeatability of the measured 
elements, some controls points have been identified along the 
track and their coordinates, determined by photogrammetric 
approach, have been compared with the “GPS-RTK” survey 
ones. 
To do that, four zones along the track have been used for 
coordinate checking. In this way it’s possible to have a good 
control of the precision reachable by this methodology. These 
four “control zones” (Figure 7) have been assigned for dividing 
all the tracks into different stretches.. The urban equipment and 
the availability of control points to be measured have been 
considered. In order to obtain the best results, the same control 
point has been measured at least in four different frames and the 
mean value of the coordinates has been used for comparison 
with the RTK value. 
Figure 7. Track of the survey overlapped on a satellite image of 
the Bologna town. The yellow square represent the four areas 
where the RTK survey has been performed. 
This method can guarantee the good quality of matching points, 
because it implies, an “a-posteriori" control on the agreement 
of the coordinates. The number of control points for each zone 
is shown in Table 2. 
Zone 
№ of CP 
№ of frames used for the mean 
1 
9 
4 
2 
15 
4 
3 
15 
5 
4 
23 
5 
Table 2. Number of Control points for each zone and number of 
frames used for the definition of the coordinate obtained by the 
Road-Scanner System