International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 200:
signal, these are replaced by other autonomous
sensors like odometer, compass, barometer, ete. Here,
we consider as the KF external measurements the
output of the photogrammetric resection; these are
the Coordinates Update (CUPT) and Attitude Update
(AUPT). In addition, Zero-Velocity Updates (ZUPT)
are also used as measurements. This kind of updates
allows adopting loosely coupled integration, which is
casiest to implement.
Since there are two cameras, two datasets of external
measurements are available; one is the EOP of the
left camera and the other is the EOP of the right
camera. There are three possibilities for this
integration.
The first possibility is to take the average of the two
EOPs. The second considers the two EOP as two
independent correlated updates. The difference
between the two possibilities is reflected in the size
and shape of the measurements’ vector, co-variance
matrix, and design matrix.
In the first possibility, the measurements’ vector can
take the following form (with ZUPTSs):
z, -la-x B-y^ C-z^ D-oqu
J
c e ©
E-owu F-emu -wv -+ ey
(16)
Where
A= OL Xon p- S Ym
2 2
C= 20 t Zon 5 + OR
2 2
EL OnR p. EL C KR
2 2
With XOL/R> YOL/R>ZoL/r = coordinates of
the left/right camera (lever-arm added)
Or;R S QLiR: KL/R = attitude of the
left/right camera (Boresight added)
c e c € S .C
X,Y 52 Op. IMU : KiMU: Vx Vy, Vz
= Position, attitude, and velocity output of
the mechanisation equations
In the second possibility, the measurements vector
can take the following form (with ZUPTS):
ZK = Kor ext Yay Zap -27 0| On
9 e
?L-9mU. KrL-Kiu. Xon -X Yon -y
€ > =
Zor —Z OR-OIMU OR-PIMU KR = KIMU
(16)
5. CONCLUSIONS AND FUTURE WORK
In this paper, we formulated a combination strategy
between photogrammetric and IMU outputs via a
Kalman Filter. We borrowed the term SLAM-
problem from Robotics and tried to solve it using
Geomatics Engineering modus operandi.
It was shown that SLAM could be achieved using
Photogrammetry and IMU outputs integrated in a
loosely coupled Kalman Filter.
The merits of this integration can be divided into
practical and scientific. Practical advantage lies in the
long-term independence of GPS. The scientific gain
poses itself in the necessity of taking
photogrammetry to the stage of full automation.
Collaboration between and merging of different
scientific disciplines —e.g. Geomatics and Robotics—
must be highly considered in order to accomplish
advancements.
Future work will concentrate on numerically testing
the methodology and the possibility of automation.
6. REFERENCES
Chaplin B. A., 1999: Motion Estimation from Stereo
Image Sequence for a Mobile Mapping System. M.
Sc. Thesis, Department of Geomatics Engineering,
University of Calgary, Canada. Report No. 20128
Csorba M., 1997: Simultaneous Localisation and
Map Building. Ph.D. Thesis Robotics Research
Group, Department of Engineering Science,
University of Oxford, UK.
Grewal M. S., Weill L. R., and Andrews A. P., 2000:
Global Positioning Systems, Inertial Navigation,
and Integration. John Wiley & Sons.
Journal of Robotics Systems: Volume 21, Issues 1
and 2, January and February 2004, Pages 1-94
http://www3.interscience.wiley.com/cei-
bin/jhome/35876. Wiley Periodicals.
Kalman R. E., 1960: A New Approach to Linear
Filtering and Prediction Problems. Journal of
Basic Engineering, 82 (series D): 35-45.
Newman P. M., 1999: On the Structure and Solution
of the Simultaneous Localisation and Map
Building Problem. Ph.D. Thesis, Australian Centre
for Field Robotics, The University of Sydney.
Australia.
Skaloud, J., Schaer, P., 2003: Towards a More
Rigorous Boresight Calibration. ISPRS
International Workshop on Theory, Technology
and Realities of Inertial/GPS/Sensor Orientation,
Commission 1, WG [/5, Castelldefels, Spain,
September 9-12.
Titterton D. H. and Weston J. L., 1997: Strapdown
Inertial Navigation Technologv. Peter Peregrinus
Ltd.
