storage
nto the
system
e-shelf
[fective
romise
urrently
Trimble
id high-
vigation
sselblad
re was
e.g., the
ometers
jPS/INS
| model,
Besides
quipped
running
form for
T
Digital
sor with
ng area),
Fairchild
ipporting
camera-
ible with
:hable to
tures an
necessary
trolled —
d 80 mm
| camera
"urations.
| and data
acquisition system around the BigShot™ camera-back
equipped Hasselblad camera involved several hardware
and software subtasks, see (Toth, 1997).
Figure 3. BigShot™ camera-back.
4. AIMS™ TEST RESULTS
AIMS™ system performance has been repeatedly
checked at different phases of the development process.
Major building blocks were first tested in laboratory
environment, and then several road tests were
conducted before the actual test flights started. Table 1
shows the major AIMS™ test flights.
Table 1. AIMS™ test flights.
Performance assessment of a navigation system is
difficult and requires an independent reference system
with accuracy characteristics at least an order better
than that of the navigation system being tested.
Unfortunately, there are no such systems available in
the AIMS™ target accuracy range. Photogrammetric
techniques can provide independently determined
position and attitude data at image exposure times.
Besides the AIMS™ camera, traditional large-format
area camera was in test flight series in order to eliminate
the camera calibration uncertainty of the AIMS™
camera. Table 2 shows typical results from test flight 7.
Photo ID eg Height difference [m]
1011 -0.09 0.22
1012 20.03 20.06
1213 0.28 20.05
1014 0.11 0.18
4041 0.09 -0.04
4042 20.31 0.10
4043 0.12 0.22
2021 0.06 0.11
2022 0.05 0.10
2023 0.15 20.20
2024 20.04 001
RMS 0.15 0.14
Table 2. AIMSTM performance data.
We believe that the evaluation results produced by the
above method might be very conservative in the sense
that real GPS/INS system performance could be much
better than evident in the evaluation results. This is
because the airborne data, as well as the ground control
simply was not good enough to produce the accurate
exterior orientation needed. Figure 4 depicts an image
taken at the Madison County Calibration Range,
London, OH, test flight 11.
Figure 4. Typical image captured over the Madison
County Calibration Range, London, OH.
5. ONGOING AIMS™ TESTS
The primary AIMS™ objectives can be separated into
two categories: (1) direct platform orientation, and (2)
promotion of digital imaging sensor technologies. It is
important to emphasize that with emerging imaging
techniques such as laser and radar imagers, platform
orientation has become an absolute necessity to achieve
proper georeferencing. In fact, there is no viablé
alternative, unlike in visual image-based techniques
where by using control points the sensor geometry can
always be reconstructed. From the very beginning, the
AIMS™ project has had a layered concept regarding the
imaging component with following phases:
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 49
