address those
very directions that will help create a highly accurate and statistically viable
result from the Patch
Test.
11.6 Basic data collection criteria
Patch test data collection does not have to be in any set order, but the order that the values
are
computed, in the data collection or processing software, will be in a distinct order.
Normally,
Latency is the first value that is computed, followed by Roll, Pitch and Yaw (or
heading). The solving
order is important, as will be seen below.
11.7 Patch Test data collection error areas
There are many common errors, or mistakes, made during the patch test data collection.
11.7.1
Positioning
The accuracy of the positioning system is a common area where errors arise. DGPS has, at best,
a
variability of ± 0.50 metres, whereas RTK variability is ±0.05metres.
A recent article, by Mike Brissette, (MosaicHydro LTD, Canada) in Hydro International (‘Stop
Using
DGPS’; Hydro International; Volume 16, Issue 7; Oct 2012) documents this issue very well:
http://www.mosaichydro.com/papers/M%20Brissette%20-%20Stop%20Using%20DGPS.pdf
The article fully details the errors that can occur by using DGPS, instead of highly accurate
positioning for the Patch Test data collection. The error increases inversely with the water
depth,
i.e. the shallower the water, the larger the error that can be induced by using DGPS over more
accurate positioning.
However, many users do not have any better positioning capabilities than DGPS; how can they
still
obtain valid patch test results without having centimetric accurate positioning? This is, in
large part,
what this paper is concerned with. However, even with centimetric position, the
following should be
followed.
11.7.2
Feature chosen for test
Where at all possible, for latency, pitch and heading data collection, a feature should be used
rather
than a slope. Slopes tend to be too variable as opposed to a well-defined feature such as
a wreck,
rock outcrop or pipeline.
One of the other issues, with using a slope, is that many times the shallow end of the slope
does not
allow sufficient area or depth for the vessel to come about and line up for the
reciprocal run; this
does not allow sufficient time for the motion sensor to settle down nor for
the helmsman to find a
steady course.
It has been found that when using a slope, for the pitch calibration, that the heading angular
offset
can have a large influence. If the sonar head does not track exactly the same route, up
and down
the slope, the heading offset will affect the pitch angular result.
Page 137 of 210
Version 5.0 Rev r002
Date 05-08-2014
To Next Page
Loading...
Need help?
General
