The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B5. Beijing 2008
865
Test
data set
SL
model
Mean
[m]
Std
[m]
Max
[m]
End
Misclosure
[m]
CEP
(50%)
[m]
Operator
S
Fuzzy
0.43
0.92
1.17
1.42
0.45
ANN
0.41
0.54
1.07
1.10
0.43
Operator
E
Fuzzy
0.59
0.43
1.25
1.14
0.59
ANN
0.62
0.47
1.11
1.26
0.65
Table 8. Statistical fit to reference trajectory of the indoor DR
trajectories generated using SL predicted with fuzzy logic and
ANN, and gyro heading; one indoor loop.
Test
data
set
SL
model
Mean
[m]
Std
[m]
Max
[m]
End
Misclosure
[m]
CEP
(50%)
[m]
327
m
Fuzzy
1.57
1.78
4.66
3.32
2.94
ANN
1.15
1.57
4.52
2.6
2.53
Table 9. Statistical fit to reference trajectory of the indoor DR
trajectories generated using SL predicted with fuzzy logic and
ANN, and gyro/compass heading; three full indoor loops.
human dynamics model. All tests to date (outdoor and indoor
environments) provided performance within the required
specifications that is below 5 m CEP50; the indoor navigation,
based on data collected to date, was limited to about 3 minutes.
More tests are underway that consider longer and more complex
indoor paths, including stairways, as this scenario has not been
tested yet. The system’s operational environment has been
originally designed for outdoor and moderately confined
environments; however, if this is to be extended to indoor
environment, additional sensors might be needed, as the human
dynamics alone may not facilitate reliable navigation for more
extended periods of time. Since the system is designed for
emergency and military crews, it cannot be expected the any
wireless infrastructure will be readily available, so the sensor of
choice should be based on imaging techniques that do not
require any additional infrastructure.
ACKNOWLEDGEMENTS
This research is supported by a 2004 National Geospatial-
Intelligence Agency NURI project.
An example outdoor trajectory, where DR solution was also
tested after a deliberate removal of the GPS signals, is
illustrated in Figure 6, and Table 10 presents the resulting
accuracy statistics.
Test
data
set
SL
model
Mean
[m]
Std
[m]
Max
[m]
End
Misclosure
[m]
CEP
(50%)
[m]
187
Fuzzy
1.74
0.93
4.14
2.19
1.46
m
ANN
2.05
1.06
4.53
3.01
1.97
Table 10. Statistical fit to reference trajectory of the outdoor DR
trajectories generated using SL predicted with fuzzy logic and
ANN, and gyro/compass heading.
T«j*ctory Reconstruction in DR
aXM
Figure 6. Reference using GPS/IMU carrier phase solution and
DR trajectory reconstructed using SL determined by FL and
ANN modules with gyro/compass heading.
REFERENCES
Bames, J., Rizos, C., Wang, J., Small, D., Voigt, G., &
Gambale, N., 2003a. Locata: A new positioning technology for
high precision indoor and outdoor positioning. Proceedings,
ION GNSS, September 9-12, Portland, OR, USA, CD ROM, pp.
1119-1128.
Bames, J., Rizos, C., Wang, J., Small, D., Voigt, G., &
Gambale, N., 2003b. High precision indoor and outdoor
positioning using LocataNet. Proceedings, Int. Symp. on
GPS/GNSS, 15-18 November, Tokyo, Japan, CD ROM, pp. 9-
18
Chiang, K., Noureldin, A., El-Sheimy, N., 2003. Multi-sensor
integration using neuron computing for land-vehicle navigation.
GPS Solutions, 6 (4), pp. 209-218.
Grejner-Brzezinska, D.A., Toth, C.K., Moafipoor, S., Jwa, Y.,
2006a. Multi-sensor personal navigator supported by human
motion dynamics model. Proceedings, 3rd IAG Symposium on
Geodesy for Geotechnical and Structural Engineering/12th FIG
Symposium on Deformation Measurements, May 22-24, Baden,
Austria, CD ROM.
Grejner-Brzezinska, D.A., Toth, C.K., Jwa, Y., and Moafipoor,
S., 2006b. Seamless and reliable personal navigator.
Proceedings, ION Technical Meeting, January 18-20, Monterey,
CA, USA, CD ROM, pp. 597-603.
Grejner-Brzezinska, D.A., Toth, C. K., Jwa, Y., Moafipoor, S.,
Kwon, J., 2006c. Multi-sensor personal navigator: system
design and calibration. Proceedings, Joint Symposium of Seoul
Metropolitan Fora & Second International Workshop on
Ubiquitous Pervasive and Internet Mapping, October 23-25,
Seoul, Korea, CD ROM, pp. 126-137.
4. SUMMARY AND CONCLUSIONS
An overview of the navigation techniques suitable for personal
navigation was presented, followed by a description of an
example implementation based on the multisensor integration
approach, using GPS, INS, magnetometer, barometer and
Grejner-Brzezinska, D.A., Toth, C.K., and Moafipoor, S., 2007a.
Adaptive knowledge-based system for personal navigation in
GPS-denied environments. Proceedings, ION National
Technical Meeting, January 22-24, San Diego, CA, USA, CD
ROM, pp.517-521.