This report treats the problem statement::

Do data collection and calculation procedures impact the accuracy of
mobile mapping data?

The answer of the above question is based on analyses and empirical
tests by means of Topcon IP-S2 HD mobile mapping and related
software.

The project is based on COWI´s ten mobile mapping tests drives on
a five kilometres long test distance south of Odense. The ten data
collections were carried out with different setups and procedures
involving the following: speed, scan sampling frequency, reference
stations and ground control points. These four parameters are
analysed based on the available dataset and software. The absolute
accuracy is checked by interpolating Z-coordinates from 71 ground
control points, surveying the route compared to the point cloud.

From the results it can be concluded that speed and scan sampling
frequency do not directly affect the accuracy of the data. Reference
distances from base to rover exceeding twenty-five kilometres entail
a considerable reduction of. By transforming the dataset´s physical
control points, it is possible to achieve an accuracy within a few
centimetres.

The overall conclusion is that at present time, without transformation,
this system is not able to deliver data with the necessary absolute
quality and accuracy which will be required for future projects.