In this paper, a newautomaticalgorithm is proposedthatfinds theoptimum locations of a Terrestrial Laser Scanner (TLS). In addition to ensuring the completeness of the data, the presented algorithm minimises the number of scanning locations. The process starts with an initial scan, and a placing a 3D grid (candidate stations) over the entire area of scan. Then a global visibility analysis is performed to identify the Next BestView(NBV) location. At this stage, the TLS is placed on the selected point and a new scan is recorded. Having updated the initial scan with the resulting point cloud, the model is checked for its completeness and density. The process is repeated until the full coverage of the scan area is reached.The selection isperformedautomatically, during which thebest globalarrangementandthe minimum number of stations needed toscan the entire area are identified. To examine the performance of the algorithm several experiments using simulated and real data were carried out. The results showed that, the algorithm is able to automatically find the stations, which can be used to coverthewholeareawith an accuracy of 99.5% (simulated data) and 91% (the real data).
