Towards Mapping of Rock Walls Using a UAV-Mounted 2D Laser Scanner in GPS Denied Environments

In geotechnical engineering, the stability of rock excavations and walls is estimated by using tools that include a map of the orientations of exposed rock faces. However, measuring these orientations by using conventional methods can be time consuming, sometimes dangerous, and is limited to regions of the exposed rock that are reachable by a human. This thesis introduces a 2D, simulated, quadcopter-based rock wall mapping algorithm for GPS denied environments such as underground mines or near high walls on surface. The proposed algorithm employs techniques from the field of robotics known as simultaneous localization and mapping (SLAM) and is a step towards 3D rock wall mapping. Not only are quadcopters agile, but they can hover. This is very useful for confined spaces such as underground or near rock walls. The quadcopter requires sensors to enable self localization and mapping in dark, confined and GPS denied environments. However, these sensors are limited by the quadcopter payload and power restrictions. Because of these restrictions, a light weight 2D laser scanner is proposed. As a first step towards a 3D mapping algorithm, this thesis proposes a simplified scenario in which a simulated 1D laser range finder and 2D IMU are mounted on a quadcopter that is moving on a plane. Because the 1D laser does not provide enough information to map the 2D world from a single measurement, many measurements are combined over the trajectory of the quadcopter. Least Squares Optimization (LSO) is used to optimize the estimated trajectory and rock face for all data collected over the length of a light. Simulation results show that the mapping algorithm developed is a good first step. It shows that by combining measurements over a trajectory, the scanned rock face can be estimated using a lower-dimensional range sensor. A swathing manoeuvre is introduced as a way to promote loop closures within a short time period, thus reducing accumulated error. Some suggestions on how to improve the algorithm are also provided.

Axis Mapping: The Estimation of Surface Orientations and its Applications in Vehicle Localization and Structural Geology

The map representation of an environment should be selected based on its intended application. For example, a geometrically accurate map describing the Euclidean space of an environment is not necessarily the best choice if only a small subset its features are required. One possible subset is the orientations of the flat surfaces in the environment, represented by a special parameterization of normal vectors called axes. Devoid of positional information, the entries of an axis map form a non-injective relationship with the flat surfaces in the environment, which results in physically distinct flat surfaces being represented by a single axis. This drastically reduces the complexity of the map, but retains important information about the environment that can be used in meaningful applications in both two and three dimensions. This thesis presents axis mapping, which is an algorithm that accurately and automatically estimates an axis map of an environment based on sensor measurements collected by a mobile platform. Furthermore, two major applications of axis maps are developed and implemented. First, the LiDAR compass is a heading estimation algorithm that compares measurements of axes with an axis map of the environment. Pairing the LiDAR compass with simple translation measurements forms the basis for an accurate two-dimensional localization algorithm. It is shown that this algorithm eliminates the growth of heading error in both indoor and outdoor environments, resulting in accurate localization over long distances. Second, in the context of geotechnical engineering, a three-dimensional axis map is called a stereonet, which is used as a tool to examine the strength and stability of a rock face. Axis mapping provides a novel approach to create accurate stereonets safely, rapidly, and inexpensively compared to established methods. The non-injective property of axis maps is leveraged to probabilistically describe the relationships between non-sequential measurements of the rock face. The automatic estimation of stereonets was tested in three separate outdoor environments. It is shown that axis mapping can accurately estimate stereonets while improving safety, requiring significantly less time and effort, and lowering costs compared to traditional and current state-of-the-art approaches.

Simulating Granite Emplacement and Deformation with Centrifuge Analogue Modelling

The Greater Himalayan leucogranites are a discontinuous suite of intrusions emplaced in a thickened crust during the Miocene southward ductile extrusion of the Himalayan metamorphic core. Melt-induced weakening is thought to have played a critical role in strain localization that facilitated the extrusion. Recent advancements in centrifuge analogue modelling techniques allow for the replication of a broader range of crustal deformation behaviors, enhancing our understanding of large hot orogens. Polydimethylsiloxane (PDMS) is commonly used in centrifuge experiments to model weak melt zones. Difficulties in handling PDMS had, until now, limited its emplacement in models prior to any deformation. A new modelling technique has been developed where PDMS is emplaced into models that have been subjected to some shortening. This technique aims to better understand the effects of melt on strain localization and potential decoupling between structural levels within an evolving orogenic system. Models are subjected to an early stage of shortening, followed by the introduction of PDMS, and then a final stage of shortening. Theoretical percentages of partial melt and their effect on rock strength are considered when adding a specific percentage of PDMS in each model. Due to the limited size of the models, only PDMS sheets of 3 mm thickness were used, which varied in length and width. Within undeformed packages, minimal surface and internal deformation occurred when PDMS is emplaced in the lower layer of the model, showing a vertical volume increase of ~20% within the package; whereas the emplacement of PDMS into the middle layer showed internal dragging of the middle laminations into the lower layer and a vertical volume increase ~30%. Emplacement of PDMS results in ~7% shortening for undeformed and deformed models. Deformed models undergo ~20% additional shortening after two rounds of deformation. Strain localization and decoupling between units occur in deformed models where the degree of deformation changes based on the amount of partial melt present. Surface deformation visible by the formation of a bulge, mode 1 extension cracks and varying surface strain ellipses varies depending if PDMS is present. Better control during emplacement is exhibited when PDMS is added into cooler models, resulting in reduced internal deformation within the middle layer.

LINEUP SIZE AND NUMBER OF CUES: WHEN BIGGER ISN’T BETTER

Larger lineups could protect innocent suspects from being misidentified; however, they can also decrease correct identifications. Bertrand (2006) investigated whether the decrease in correct identifications could be prevented by adding more cues, in the form of additional views of lineup members’ faces, to the lineup. Adding these cues was successful to an extent. The current series of studies attempted to replicate Bertrand’s (2006) findings while addressing some methodological issues—namely, the inconsistency in image size as lineup size increased. First, I investigated whether image size could affect face recognition (Chapter 2) and found it could, but that it also affected previously-seen (“old”) versus previously-unseen (“new”) faces differently. Specifically, smaller image sizes at exposure lowered accuracy for old faces, while these same image sizes at recognition lowered accuracy for new faces. Although these results indicate that target recognition would be unaffected by image size at recognition (i.e., during a lineup), lineups are also comprised of previously-unseen faces, in the form of fillers and innocent suspects. Because image size could affect lineup decisions, as it could become more difficult to realize fillers are previously-unseen, I decided to replicate Bertrand (2006) while keeping image size constant in Chapters 3 (simultaneous lineups) and 4 (simultaneous-presentation, sequential decisions). In both Chapters, the integral findings were the same: correct identification rates decreased as lineup size increased from 6- to 24-person lineups, but adding cues had no effect. The inability to replicate Bertrand (2006) could mean that the original finding was due to chance, but alternate explanations also exist, such as the overall size of the array, the degree to which additional cues overlap, and the length of the target exposure. These alternate explanations, along with directions for future research, are discussed in the following Chapters.