rder to produce
ence, indicating
range measur-
ield of view.
the laser range
iation of range
ormal ambient
r a heating up
of the scanner
outside heater,
Was detected.
th the scanner
during an out-
€).
ner
rer, or its re-
and a series of
tic point in the
ition in each of
of a small wall
range data and
“jump edges of
e scanner is not
o the scanning
er profile, only
about 8 mm/m,
ection
ns of the laser
uated. Figure 6
be measured in
tunnel tube are
system. Tunnel
faces) is in the
tunnel
m
00 3000
4000
3500
3000-2] ct
polar range [mm]
2010 eMe MM
2000 [- db ll
i | i | |
50 100 150 200 250 300
scanning angle ß [deg]
6b) Polar plot of reference tunnel profile
Fig. 6: Reference tunnel profile
Figure 7 shows the profile measured with the laser range scan-
ner and compares it to the reference tunnel.
4000 --—-
3500
w
©
co
e
T
polar range [mm]
210 EeÁJ
| i | i | |
50 100 150 200 250 300
scanning angle ß [deg]
7a) Measured profile (polar plot)
©
ce O1
T
log10 of range error [mm]
e
N
| | ] | | |
“50 100 150 200 250 300
scanning angle b [deg]
7b) Deviations with profile measurement
Fig. 7: Tunnel profile measurement results
Resulting deviations between reference and measured profile
of several mm correspond to examinations of Chapter 3 and 4
and fulfill all demands with tunnel surface inspection. Corres-
ponding grey level profile is used for visual inspection, surface
classification, and documentation purposes.
5. CONCLUSIONS
In this paper we have examined in detail the measurements
supplied by an amplitude-modulated laser range sensor. We
presented experimental performance data and discussed the ap-
plication of the sensor hardware for tunnel inspection. The
quantitative performance, in terms of accuracy and precision
with different surface material types, fulfills the demands of
tunnel surface inspection. Compared [20] to other laser range
sensors, the laser range measuring system shows good perfor-
mance data. Robustness of the laser range scanner even under
varying temperature and lighting conditions is sufficient for
application in industrial environments. Ambiguity interval of
range measurement and range accuracy of the laser range scan-
ner may be adapted to other applications by selecting other
modulation frequencies.
Parallel to "range" measurements reflectance of surface ma-
terial "active grey level", is measured. Both, range and grey
level data of a target point are registered at the same time and
correspond to one single target point defined by the laser beam
direction. The range profile of the laser range scanner reflects
geometric dimensions of the tunnel tube whereas the grey
level profile is used for visual inspection, surface classifica-
tion, and documentation purposes.
In a next step of our research we will test the laser range
scanner in several railway tunnels and mount it onto a vehicle
in order to achieve range and grey-level images of the com-
plete tunnel tube. Effects of ambient temperature gradients in
the tunnel as well as humidity in the tunnel to range accuracy
and precision have to be studied in detail.
ACKNOWLEDGEMENTS
The work reported in this paper was supported by the
Deutsche Forschungsgemeinschaft, as part of an interdiscipli-
nary research project on "Information processing in Autono-
mous Mobile Robots" (SFB 331) and by Spacetec GmbH, Frei-
burg, Germany. The authors would thank Dipl. -Ing. (FH) R.
Wigler (Spacetec GmbH) for the improvements of the receiver
optics.
REFERENCES
[1] Schmidt, G.: "Towards Integration of Autonomous Sub-
systems for Assembly and Mobility into Flexible Manu-
facturing". Proc. of Internat. Workshop on Information
Processing in Autonomous Mobile Robots, München,
(1991), pp. 3 - 20
[2] Besl, P.: "Active, optical range imaging sensors". Ma-
chine Vision & Applications. Vol. 1, (1988), pp. 127-152
[3] Kanade, T.: Three-Dimensional Machine Vision. Kluwer
Academic Publishers, 1987
477