Fig.4 - TAG and batteries supply installation for van
experiment.
Fig.5 — LN-200 IMU mounted on to the IMU-Bench prototype.
4.2.1 Data Analysis
Once the route was finished and the data collected, the RINEX
and raw IMU measurements were processed in order to
compare them with the real time solution that was also stored in
the TAG. Some of the results of the comparisons can be seen in
figures 6, 7, 8 and 9.
PROCESSED OUTPUT TRACK — 1
p —
Viladecans
a
eves amet
Castelldefels Beach
1.96 1.98 2.00 2.02 2.04
Lon(deg)
Fig.6 — Route run by the car in the experiment
The comparison of the trajectories shows that the difference in
position between the two navigation solutions may vary to a
maximum of about 15 meters (Fig.7). This is due to different
quality of the GPS positions used in real time and in post-
processing. The measurements took place in urban and sub-
urban areas, where the conditions for GPS positioning are not
optimal (satellite visibility, multipath). It is assumed, that the
post-processing solution for the GPS trajectory is more precise.
So, given that the GPS trajectory is an important part of the
navigation processing, the cause of difference in position is
known.
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part Bl. Istanbul 2004
~ END POINT
1.97 1.57 un , 1.084
Fig.7- Zoom of experiment route. In Red it is marked the real
time solution, for comparison purposes.
For the attitude angles the performance in part is good. The
maximum difference between the real time and the post-
processed pitch and roll is about 0.2 degrees (Fig.9). The
heading varies about | degree, which is again due to the
imprecision of the real time GPS trajectory when working in
urban environments. That trajectory is internally used by the
navigator for self-calibrating of the heading gyro.
290F T T
r4 y
Fig.8 - Real time solution of heading angle.
j
v des. v À * PM Nf Ss ORE, eunt 1
& : A : s i
i V |
|
rv 4
v se LA
manette, i
ms A ng |
È 1
E i
E 1
È j
|
Fig.9 - Comparison of attitude angles between real time
solution and post-processed solution. The black line
is heading, the red one is pitch and the blue one is
roll.
Only a short analysis of the results could be carried out for this
paper. We will conduct a more detailed analysis in the close
future.
4.4 Future experiments
In addtion to this, TAG is also being used in GeoPIE project,
where it is helping to develop the navigation model of a remote
sensor. In this case, two flights have been carried out with the
TA
dev
Firs
AS
per
new
raw
and
hav
[MI
sup
It 1
syst
Wit
Sysl
nav
qua
wid
app
of 1
gra
of (
the
Wh
dev
dev
Bai
Tin