The accuracy of measuring distances with the VISAT system is
another way for testing the system's relative accuracy. For this
purpose a special test was done by measuring the length of some
well defined features along the test course and taking some
images for these features in kinematic mode. Figure (10) shows
the difference between the VISAT computed distances and the
known distances. The figure shows that these errors are distance
dependent, as expected, and reach a magnitude of 10-14 cm for
objects 30 m away from the van.
15
| d
zo upuooo noiteie ierit e
E * + e v
2 5 e e? e
8 rà: ee eee
8 oe ®
= 0 T T T T T T T T T T
#15 | Bail 200 t | 15081-12000
8 e e .
maglia DLL... N.S
-15 2
02 4 6 8 1012 14 16 18 20 22 24 26 28 30
Distance From the Van (m)
Figure 10 . The Accuracy of Measuring Distances With
The VISAT System
4.2 Absolute Accuracy
The main objective of the VISAT system is the determination of
the 3-D coordinates for all objects within the video cameras' field
of view. The final accuracy of the 3-D coordinates is a function
of the complete processing chain which involves GPS positions,
INS position/attitude, target localization in the images, and
system calibration. Figure 11 shows the errors in the computed
3-D coordinates of 14 control points located along one of the test
sectors. The 3-D coordinates are computed from 2-images using
the georeferencing formula described in Section 2. The distance
between the control points and the cameras was approximately
10-30 m. The figure shows clearly that an RMS of 16 cm in the
horizontal coordinates and 7 cm in height are achievable for
distances up to 30 m away from the van.
CONCLUSION
The VISAT system presented in this paper is a mobile multi-
sensor system which can be operated continuously under many
operational conditions. In particular, the complementing features
of the integrated INS/GPS positioning and orientation subsystem
permit the resolution of cycle slips and outages as well as INS
drift control. System calibration results indicate that
accuracies of 7 cm (RMS) can be obtained. System testing
indicate that highway velocities of 60 km/h can be maintained
with adequate data transfer and target positioning in a post-
processing mode at a workstation. Run-to-run and day-to-day
repeatability achieved in the first system testing is about 8 cm
(RMS) in horizontal and about 6 cm (RMS) in height. There is
no significant difference between run-to-run and day-to-day
results. This indicates that the system positioning component,
GPS/INS, works at a consistent level from day to day. In
general, the results of system testing show that the absolute
accuracy surpasses the development objectives for the system..
30
e
UT ai TEA RU re
5 jo 4... 3... di as tee . 2 reel, a eru !
A
2 à A A amas "i
- 0 T T T T T T
E " à
A
Boye der A BE BE
2-10 = 8 s» ea^ ie
& e e
2098 SE VE
e
-30
9-2 «6.0. 42M
| eE sN ah | Point Number
Figure 11. The VISAT System Absolute Accuracy
ACKNOWLEDGMENTS
The VISAT system represents the combined effort of a research
team at the Department of Geomatics Engineering of the
University of Calgary and Geofit Inc., Laval, Quebec.
REFERENCES
[1] Cosandier, D., Chapman, M. A, High Precision
Target Location for Industrial Metrology, Videometrics,
SPIE OE/Technology, Boston, November, 1992.
[2] El-Sheimy, N., Schwarz K.P., Integrating Differential GPS
Receivers with an Inertial Navigation System (INS) and CCD
Cameras for a Mobile GIS Data Collection System, ISPRS94,
Ottawa, Canada, October, 1994, pp. 241-248.
[3] El-Sheimy, N., Schwarz K.P., and Gravel M., Mobile 3-D
Positioning Using GPS/INS/Video Cameras, The Mobile
Mapping Symposium, Ohio State, USA, May 24-26, 1995, pp.
236-249.
[4] Li, R., M. A. Chapman, Qian, L., Xin. Y., and K. P. Schwarz,
Rapid GIS Database Generation Using GPS/INS Controlled
CCD Images, ISPRS 94 GIS/SIG, June 6-10, Ottawa, Canada.
[5] Schwarz, K. P., Martell, H., El-Sheimy, N., Li, R., Chapman,
M., Cosandier, D. (1993a): VISAT- A Mobile Highway Survey
System of High Accuracy, VNIS Conference ‘93 Conference,
Ottawa, October 12-15, 1993, pp. 476-481.
[6] Schwarz, K.P., Chapman, M.A, Cannon, M. W., Gong, P.
(1993b): An Integrated INS/GPS Approach to the
Georeferencing of Remotely Sensed Data, PE&RS Vol. 59, No.
11, November 1993, pp. 1667-1674.
100
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B2. Vienna 1996
im
ch:
