3. AIM OF THE EXPERIMENTS 
As mentioned in the introduction, one of the aims of the 
experiments was to determine, with the GPS, the coordinates of 
a sufficient number of points to use for the merging of scans 
performed with the laser scanner. 
Merging of the various point clouds is based on a 
rototranslation in space. This transforms the coordinates of all 
the points of the cloud (each scan is referred to a "local" frame 
centred on the laser scanner) into a unique or "global" frame 
defined by the user, which can coincide with the local frame of 
a predefined scan. 
Therefore, for each scan, six parameters of the transformation 
must be calculated (the three rotations of the axes and the 
components of the vector of translation of the origin of the local 
frame); using three homologous points and a least squares 
procedure, it is possible to calculate the parameters of the 
rototranslation with a degree of superabundance of three; thus it 
is possible to apply the transformation to the entire point cloud. 
Normally the homologous points allowing an accurate recording 
between two or more scans are constituted by flat reflecting 
targets positioned on the object to be measured; the scans to be 
joined must have a large area of overlap and the position of the 
target must be analysed from time to time, making the 
surveying of large territories very difficult (hereafter we use the 
term territorial scale surveying). 
The idea used in these experiments was to successively perform 
all the scans necessary to cover the survey area and, for each of 
them, to position at least three GPS antennas able to furnish the 
coordinates of three homologous points (between the GPS and 
laser scanner system) to be used to calculate the parameters of 
the transformation. In this way, the WGS84 geocentric system 
was used directly as a "global frame" for all the scans. 
Naturally, the GPS antenna, more precisely its phase centre, 
could not be used directly as a homologous point (measured by 
both the GPS and laser scanner); it was necessary to make a 
suitable device to allow the simultaneous positioning of the 
GPS antenna and a classical flat reflecting target. In this way, 
the GPS measurement furnishes the WGS84 coordinates of the 
antenna’s phase centre, while the laser scanner provides the 
coordinates of the centre of the reflecting target in the local 
frame of the scan. At this point, the WGS84 coordinates of the 
antenna’s phase centre must be transferred to the target centre, 
an operation requiring an initial control and calibration of the 
adapter devices (described later). 
The advantages of this solution are obvious: 
contemporaneous measurements', the time component is an 
important factor in territorial scale surveying; 
- flexibility of use of the instrumentation: the proposed 
method satisfies the requirements of the single instruments 
without creating annoying overlaps of one on the other; 
areas of overlap reduced to a minimum', with the proposed 
method, no overlap between the different scans is required 
for the recording; 
planning of the surveying directly in the field', exploiting the 
operational elasticity of the GPS antennas, it is possible to 
decide on the spot which portion of the territory is to be 
scanned and where to position the reference antennas, 
without having to plan the session in the laboratory. 
3.1 Planning and calibration of the adapter devices 
To achieve the described aims, we designed and built a suitable 
adapter equipped with a Wild fillet that allowed us to 
contemporaneously position both the flat reflecting target and 
the GPS antenna on a topographical tripod and a sideboard 
equipped with a spherical level. 
a », 
a « 
Figure 4. Design of the adapter 
To contemporaneously perform the GPS and laser scanner 
measurements, we equipped the plane holding the reflecting 
target with two orthogonal axes of rotation that allowed us to 
maintain it in a frontal position with respect to the scanner; in 
this way, automatic target recognition was possible in the 
scanning phase and manual recognition was facilitated at great 
distances. 
Figure 5. The adapter we built 
For each adapter, we first performed a control procedure to 
verify the correctness of its construction. In particular, we 
measured the planimetric difference in the position of the target 
centre with respect to the vertical for the station point and the 
change in position of this centre when the adapter was rotated 
around the vertical. 
In addition, we performed a careful calibration that allowed us 
to assess the vertical offset between the coordinates of the phase 
centre of the GPS antenna and the coordinates of the centre of 
the flat reflecting target positioned on the adapter. 
Figure 6. The vertical offset between the phase centre and the 
target’s one 
-29-