Systematic Influences on geodetic Totalstation-Measurements
Within geodetic Monitoring-Applications, the used sensors will very often have to be protected against harsh weather and unwanted access of unauthorized persons. As an example for such a protective housing, the one used for a monitoring project operated by the Engineering Geology Group of ETH-Zürich at the Aletsch-Glacier in the canton of Valais, Switzerland can serve.
It is a known fact that every obstructions in, or close to the line of sight might influence angular measurements as well as distance measurement, but the magnitude of error encountered with this particular type of housing (up to 20mgon at sighting distances of app. 600m) could not be explained by single error sources alone.
In order to determine the sources of influence, and provide a solution that meets accuracy as well as protection demands, a small scale laboratory experiment was set up.
By automatically rotating the protective housing around a Total-Station that performs set-measurements to a stable set of targets, one can determine the influence for every possible line of sight.
By simple Ray-Tracing simulations it could be shown that the effects on angular measurements as well as distance measurement are caused by the effects of the windows represented as optical wedges. These optical wedges introduce errors depending on the angle of incident at the window and the length of the optical path within it. Combining this with the experimentally found influence onto the ATR, the observed errors for a line of sight passing close to a non-transparent obstruction can easily be explained.
For protective housings of any construction, that incorporate windows and other obstructions within or close to the line of sight these sources of error can never be fully eradicated. So an alternative solution had to be found that causes as less influence on the measurements, and still offers as much protection as possible. A simple solution was found to be a commercially available product where after installation on the site, it only is necessary to drill a hole into a plastic cylinder at each line of sight to be monitored.
As the plastic material is non-transparent, the distance measurement will not be affected by differences in the optical path length, even if the edges of the hole would be visible to the instrument. Due to the effects on angular measurements caused by obstructions visible for the ATR-System it is important to use a hole diameter that is small enough to still offer protection, but big enough to not affect the ATR when taking into account point-movement and misalignment of the protective cylinder, for example after maintenance of the system.
In order to determine the hole-size best to be used a strictly experimental approach was chosen, where by rotating the holes with respect to the line of sight axis misalignment could be tested.
The results show, that with line of sight and axis of the hole coaxial, which means that the disturbance seen by the ATR is symmetrical, unaffected measurements can be achieved for all tested diameters. With increasing misalignment of the axis the errors increase rapidly. For a hole-diameter of app. 31 mm a misalignment of just 0.25 gon will cause an effect that already exceeds the instruments accuracy as stated by the manufacturer. With a hole-diameter of 45 mm the misalignment necessary to cause the same effect can be about 2 gon what will in most cases offer enough margin for point-movement and misalignment.
Contact
I. f. Geodäsie u. Photogrammetrie
Stefano-Franscini-Platz 5
8093
Zürich
Switzerland