HR: 0800h
AN: G51B-0087 [Abstracts]
TI: Reconstructing 3-D Ship Motion for Synthetic Aperture Sonar Processing
AU: * Thomsen, D R
EM: dthomsen@ucsd.edu
AF: Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla, CA 92093-0225
United States
AU: Chadwell, C D
EM: cchadwell@ucsd.edu
AF: Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla, CA 92093-0225
United States
AU: Sandwell, D
EM: dsandwell@ucsd.edu
AF: Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla, CA 92093-0225
United States
AB:
We are investigating the feasibility of coherent ping-to-ping processing of multibeam sonar data for high-resolution mapping
and change detection in the deep ocean. Theoretical calculations suggest that standard multibeam resolution can be improved
from 100 m to ~10 m through coherent summation of pings similar to synthetic aperture radar image formation. A requirement
for coherent summation of pings is to correct the phase of the return echoes to an accuracy of ~3 cm at a sampling rate of
~10 Hz. In September of 2003, we conducted a seagoing experiment aboard R/V Revelle to test these ideas. Three
geodetic-quality GPS receivers were deployed to recover 3-D ship motion to an accuracy of +- 3cm at a 1 Hz sampling rate
[Chadwell and Bock, GRL, 2001]. Additionally, inertial navigation data (INS) from fiber-optic gyroscopes and pendulum-type
accelerometers were collected at a 10 Hz rate. Independent measurements of ship orientation (yaw, pitch, and roll) from the
GPS and INS show agreement to an RMS accuracy of better than 0.1 degree. Because inertial navigation hardware is susceptible
to drift, these measurements were combined with the GPS to achieve both high accuracy and high sampling rate. To preserve
the short-timescale accuracy of the INS and the long-timescale accuracy of the GPS measurements, time-filtered differences
between the GPS and INS were subtracted from the INS integrated linear velocities. An optimal filter length of 25 s was
chosen to force the RMS difference between the GPS and the integrated INS to be on the order of the accuracy of the GPS
measurements. This analysis provides an upper bound on 3-D ship motion accuracy. Additionally, errors in the attitude can
translate to the projections of motion for individual hydrophones. With lever arms on the order of 5m, these errors will
likely be ~1mm. Based on these analyses, we expect to achieve the 3-cm accuracy requirement. Using full-resolution
hydrophone data collected by a SIMRAD EM/120 echo sounder we are applying the 6 components of ship motion to correct the
phase center of each hydrophone. Successive pings will be analyzed for phase coherence.
DE: 3045 Seafloor morphology and bottom photography
DE: 3094 Instruments and techniques
DE: 3099 General or miscellaneous
DE: 1294 Instruments and techniques
DE: 1299 General or miscellaneous
SC: Geodesy [G]
MN: 2004 AGU Fall Meeting