AnsweredAssumed Answered

SX10 Scanning Range Error Filter

Question asked by John Whidden on Jul 9, 2020
Latest reply on Jul 10, 2020 by John Whidden

The new Trimble SX10 Scanning Total Station has a published scanning range
error of 1.5mm +/-. The actual Range error is 1.5mm +/- at one sigma, or in real terms 4.5mm
+/-. (see fig. 1 attached)
For earthworks and many civil applications, the range error of 4.5mm +/- is of minimal
consequence and is hardly noticeable when surveying earthworks, road corridors, quarry
surveys etc…
However, when completing higher order surveys for feature deformation, monitoring, as-built
of new construction, etc… where construction survey tolerances are less than 5mm, this
becomes problematic. There are also concerns when scanning detailed features that are to be
meshed, the range error creates a rough result.
Though the Trimble SX10 Scanning Total Station produces a point cloud with an
overall thickness of 8-10mm, the mean of the point cloud yields a result of less than 1mm +/-
variation when compared to Total Station Direct Reflect observations.
Trimble Business Center currently has a Best Fit Line and Best Fit Arc routine
that will fit lines to the mean of a cloud section, but to the best of my knowledge there no
routines to filter and or reduce the point cloud to the mean of the point cloud, or to create a
surface mesh based on the mean of the point cloud.


Requested Features
1. Step 1 - The ability to filter the point cloud based on the mean of the point cloud. For
example: there would be a user variable to set the desired depth of points to retain
relative to the mean of the point cloud. (see fig.2 & fig.3 attached) If that variable was
set to 3mm, the resulting point cloud would have an overall thickness of 6mm and all
outliers would be removed.
2. Step 2- The ability to reposition the resulting points from step 1 to fit the mean of the
point cloud, all points remaining from step 1 would have their slope distances adjusted
relative to the scanner position so that they are repopulated at the mean of the point
cloud. (see fig.4 attached)