The purpose of this paper is to investigate the height uncertainty of digital terrain models (DTMs) generated from unmanned aircraft system (UAS) surveys over different surface types – asphalt, gravel and grass. The data used in the study was acquired during two UAS flights performed in spring 2014 with a fixed wing aircraft and two different cameras, from the flying height of about 100 m, and it was processed in different software suites – Agisoft PhotoScan, RapidStation and RapidTerrain. The results show that it is possible to achieve the height uncertainty (expressed as a Root Mean Square Error) in a DTM of below 0.02 m on asphalt surfaces and below 0.04 m on gravel and grass surfaces, provided an overcast sky.

In order to investigate the possibilities for the generation of digital elevation models with a height uncertainty of 20 mm or less using unmanned aerial vehicles, a survey with a fixed-wing aircraft was carried out over a gravel quarry in Gävle, Sweden. At flying heights of about 80 and 160 m, the resulting imagery had ground sample distances of 24 and 50 mm, respectively. The data was processed in two different software suites – computer-vision-based PhotoScan and photogrammetric RapidStation/RapidTerrain. The results show that PhotoScan was more effective in flat terrain, with height uncertainties of 10 and 17 mm compared with 30 to 40 mm for RapidTerrain, whereas the latter had a clear advantage over undulating terrain.

Syftet med det här projektet har varit att undersöka utbudet av befintliga UAS för laserskanning, studera erfarenheter av deras tillämpning inom 3D-kartering, samt bedöma deras potential för framställning av DTM för byggnadsinformationsmodellering (BIM). Med tanke på teknikens snabba utveckling bör den här rapporten ses mer eller mindre som en ”ögonblicksbild” av ”UAS-laserskanningslandskapet” i slutet av 2015.

Byggnadsinformationsmodellering (BIM) är ett begrepp som avser att ge en effektiv byggprocess genom digital representation av både fysiska och funktionella aktiviteter i ett byggprojekt. I en vid bemärkelse kan BIM användas i hela byggprocessen, från planering, projektering och byggande till förvaltning, renovering och slutlig destruktion. I en mer snäv bemärkelse används BIM enbart för digitala modeller (Volk m.fl., 2013), som i den här rapporten, där fokus ligger på framställning av digitala terrängmodeller (DTM) för exempelvis byggprocessens planering och/eller projektering.

Methane built up in landlls as a result of breaking down of organic materials can be a renewable energy source if it is taken advantage of. The aim of research presented in this paper is to detect landll gas (that contains methane) by means of terrestrial laser scanning. The hypothesis is that where no surface leakage has been reported, the landll gas will expand or migrate. Therefore, it is possible to detect it through repeated scanning of the same area and comparison of Digital Terrain Models (DTMs) generated from the point clouds. Only the most signicant movements, i.e. vertical, are of interest in this case. During September–November 2011, a small area at Forsbacka landll in the vicinity of Gävle was scanned 10 times. Epoch-to-epoch comparisons of the resulting DTMs have shown two signicant changes (–27 and +19 mm) in elevation of the surface, and it is not impossible that they are caused by migrating landll gas. The method tested in this study is deemed to be rigorous and accurate for detecting small-scale swell-shrink behaviour of the ground surface (in our case a landll surface). However, both data processing and interpretation of the results have been considerably complicated by presence of low vegetation (weeds) on the study site, which was di-cult to lter away completely from the data. Based on our pilot study, we recommend that a larger area and a longer period of time are chosen to give basis for more grounded conclusions about presence of landll gas.

På Högskolan i Gävle har under vårterminen 2013 fyra examensarbeten genomförts för att undersöka möjligheter och problem med användandet av obemannade flygfarkoster för fotogrammetriska ändamål. Studenterna har läst det treåriga lantmätarprogrammet med teknisk inriktning, som liksom alla högskoleprogram, avslutas med ett examensarbete. De här fyra grupperna har haft ett gemensamt intresse, nämligen UAS, och deras undersökningar refereras i artikeln nedan. Men först en kort beskrivning av UAS.

This study uses results from a series of analogue models, and field observations, scanned data and sections of natural landslides to investigate the kinematics and internal deformation during the failure of an unstable slope. The models simulate collapse of granular slopes and focus on the spatial and temporal distribution of their internal structures. Using a series of systematically designed models, we have studied the effect of friction and deformability of the runout base on internal deformation within a granular slope. The results of these different models show that the collapse of granular slopes resulted in different-generation extensional faults at the back of the slope, and contractional structures (overturned folds, sheath folds and thrusts) at the toe of the slope. The failure surfaces and the volume of the failure mass changed both spatially and temporally. Younger failure surfaces formed in the back of the older ones by incorporating additional new material from the head of the slope. Our model results also show that the nature of the runout base has a significant influence on the runout distance, topography and internal deformation of a granular slope. Model results are compared with natural landslides where local profiles were dug in order to decipher the internal structures of the failure mass. The natural cases show similar structural distribution at the head and toe of the failure mass. As in model results, our field observations indicate the presence of at least two generations of failure surfaces where the older ones are steeper.

The collapses of granular materials frequently occur in nature in the form of, for example, rock avalanches, debrisavalanches and debris flow. In previous studies of collapses of a granular material, most of the focus has been onthe effect of initial geometry and mechanical properties of the granular materials, the run-out distance, and thetopography of final deposit. In this study, results of analogue models and scanned natural failed slopes are usedto outline the mode of failure of an unstable slope. Model results and field observations are used to argue that agranular mass moves downslope in a wavy pattern resulting in its intensive deformation.In the models, we mainly investigated the internal deformation of collapses of granular slopes in terms of theirinternal structures and the spatial and temporal distribution of the latter. Model results showed that a displacedmass of the granular slope has the following two features: (1) Initial collapse resulted in a series of normal faults,where hanging-wall blocks were slightly deformed, like the slump-shear structures in nature; (2) With furthercollapse, a set of secondary structures, such as deformed/folded fault surfaces, faulted folds, displaced inclinedfolds, and overturned folds formed near the slope surface. The occurrence of these structures reflects the failureprocess of the granular mass in space and time. In addition, our model results show that the nature of basal frictionhas a significant influence on the geometry and kinematics of these structures at the slope toe. Model results showalso that the mass does not glide downslope along only one surface, but includes several gliding surfaces each ofwhich take part of the sliding. These gliding surfaces become steeper deeper in the sliding mass. Some of thesefeatures observed in the models are also detected in the field. Scanned failed slope surfaces show a wavy patternsimilar to that in the models, reflecting the presence of normal faults at the head of the slope and folding at theslope toe.

This paper reports on investigation of measurement uncertainty in positioning of a terrestrial laser scanner with network RTK (Real-Time Kinematic) service provided by SWEPOS®, Swedish national network of permanent reference stations for GNSS (Global Navigation Satellite System). To simulate measurements by a scanner, a rotating flat bar fixed to a prism base, attached to a tribrach, was used. The tests have been carried out with both a rotating GNSS antenna (placed at different distances from the centre of rotation – radii) and a stationary antenna, under different time intervals (1–5 min). The results show that it is possible to achieve a standard uncertainty of less than 10 mm in plane and 16 mm in height, independently of the observation time and radius. Hence, network RTK can be used with advantage for precise direct georeferencing of point clouds, not only for determination of the position of the scanner, but also its orientation.