Global Positioning System Reference
In-Depth Information
resolution is 0.25°; that is, the density of 3D laser points is approximately 20 cm per point.
After comparing the mapping accuracy with the laser point density, it was found that the
accuracy is good enough for mapping DSM. Therefore, the accuracy of hybrid positioning
including their attitude is considered approximately 10 to 30 cm.
Unit: m
Ground Control point
No.
(X)
(Y)
(Z)
(X)
(Y)
(Z)
(X)
(Y)
(Z)
1
-11184.877
-25630.253
42.755
-11184.696
-25630.836
42.915
0.181
0.583
0.160
2
-11185.471
-25622.727
42.952
-11185.557
-25622.789
42.971
0.086
0.062
0.019
3
-11167.603
-25670.474
42.391
-11168.282
-25670.312
42.406
0.679
0.162
0.015
4
-11177.107
-25634.721
42.704
-11177.262
-25634.918
42.523
0.155
0.197
0.181
5
-11152.866
-25641.753
42.029
-11152.172
-25641.036
42.071
0.694
0.717
0.042
6
-11176.511
-25625.571
42.824
-11176.467
-25625.426
42.767
0.044
0.145
0.057
7
-11153.911
-25643.823
42.534
-11154.375
-25643.041
42.075
0.464
0.782
0.459
8
-11150.564
-25631.724
42.340
-11150.887
-25631.869
42.296
0.323
0.145
0.044
9
-11176.771
-25635.344
43.992
-11176.394
-25635.308
44.082
0.377
0.036
0.090
10
-11186.666
-25631.657
44.289
-11186.417
-25631.888
44.202
0.249
0.231
0.087
DSM DSM DSM
Error Error Error
Av e . E r ro r
0.325
0.306
0.115
Table 5. Positioning accuracy assessment from DSM
7. Conclusion
In this paper, robust trajectory tracking by hybrid positioning was developed and a digital
surface model was reconstructed with multi-sensor integration using entirely inexpensive
sensors, such as a small laser scanner, digital cameras, an inexpensive IMU, a GPS, and a
VMS. A new method of direct geo-referencing was proposed for laser range data and
images by combining a Kalman filter and the BBA or VMS. Because the result of BBA avoids
the accumulation of drift errors in the Kalman filtering, the geo-referenced laser range data
and the images were automatically overlapped properly in the common world coordinate
system. Hybrid positioning data is acquired by using or combining several different
positioning technologies. Since this paper focused on how to integrate the sensors into a
mobile platform, all the sensors and instruments were assembled and mounted under a
mobile platform such as a UAV or ground vehicle in this experiment. Finally, the precise
trajectory, including attitude of the sensors, was computed as the hybrid positioning for
direct geo-referencing of a laser scanner. The hybrid positioning data is used to reconstruct
digital surface models.
8. References
Zhao, H. & Shibasaki, R. (2000).
Reconstruction of Textured Urban 3D Model by Ground-Based
Laser Range and CCD Images
, IEICE Trans. Inf.&Syst., vol.E83-D, No.7
Manandhar, D. & Shibasaki, R. (2002).
Auto-Extraction of Urban Features from Vehicle-Borne
Laser Data
, ISPRS, GeoSpatial Theory, Processing and Application, Ottawa
Search WWH ::
Custom Search