1978 ISP COMMISSION V INTER-CONGRESS SYMPOSIUM - STOCKHOLM SWEDEN
— — — — "am a A— "m "-— -— "aX -— enn uet eene "hes Hem Seen Hee amm bm damn unum D Mene vete weem Wein vem em uo m m v id ns
— a9À— — «—" — — r— "-— unum emp emm nat quem m mem uem m wee aun wem o me m m m n m — A ne — — —K —
Eales: up the instrument. You have to do your relative and absolute orien-
tations before you can start to take measurements to points on the cross
sections. Now once you have got the relative orientation set-up the obser-
vations of the individual cross sections can be very rapid. Therefore if
we have only a comparatively few points to measure you can use an alterna-
tive technique which perhaps takes longer to measure each individual point,
but does not require the long set-up procedure before you can start to make
the measurements. In this case we felt that our method was a lot faster
than the photogrammetric solution.
Konecny: I think from the technical point of view we agree. On the other
hand it would be interesting to note why you only need seven points.
Fales: A train is essentially a square box construction. Most tunnel cross
Sections are circular. Therefore the critical points tend to be the four
corners of the rolling stock. Therefore those four corners were observed
in the tunnel cross section, and in addition three other points in which
the engineers were interested, the crown and the two side axes of the tunnel.
That was all he was interested in in order to determine whether he could
get his train through the tunnel.
Konecny: I think there are interesting analogies in DIM's and highway
engineering cross sections.
Newton: Could you tell us to what extent the photogrammetric system set up
by one of the consortiums building oil production platforms in the United
Kingdom is currently being used?
Eales: The answer is that it is not being used very much. I think this is
a reflection of the extreme cost-effectiveness of the technique that was
adopted. If we consider the history of the situation, we started off with
knowing that a number of the subassemblies were being fabricated outside the
required tolerances. The next requirement was therefore to determine where
these deviations were, so that the fabrication sequence could be amended to
cope with these deviations. For this a very good measuring system was re-
quired. And this is where the photogrammetric solution came in. Now, how-
ever, partly perhaps through the improved consruction techniques, and also
because we got a feedback of information as to the nature of the deviations
of these components, the subassemblies are now being fabricated to much
higher tolerances, and we have found that only a few simple check measure-
ments are normally required to satisfy the engineer that the component that
has been fabricated is within tolerances. Therefore the cost-effectiveness
is that we are not have to pay for amendments to the fabrication technique,
in spite of the fact that the photogrammetric unit is not being used.
Atkinson: I was personally involved in the photogrammetric Wriggle Survey
which Dr. Eales referred to in his paper. A comment from my experience on
the Mersey tunnel work with regard to the specific points which Dr. Eales
made about control, lighting, and targetting. There was a similarity of
requirement in terms of the number of points per cross section. The length
of the tunnel which was covered was 50 meters with cross sections at 8 foot
intervals, so that we were talking roughly 160 points to be co-ordinated
from one pair of pictures. The point there is that, from one pair of photo-
graphs a substantial number of points could be co-ordinated to the required
accuracy. And the number of control points for that one model was four.
The lighting was a problem in that experimental work. We suggested certain
refinements. Whether they would have been successful I do not know. One of
