Gegenstand dieser Arbeit ist die Untersuchung von statistischen Eigenschaften der ungeordneten Metallen im Rahmen des Anderson-Modells der Lokalisierung. Betrachtet wird ein Elektron auf einem Gitter mit "Nächste-Nachbarn-Hüpfen" und zufälligen potentiellen Gitterplatzenergien. Wegen der Zufälligkeit zeigen die Elektroneigenschaften, zum Beispiel die Eigenenergien und -zustände, irreguläre Fluktuationen, deren Statistik von der Amplitude der Potentialenergie abhängt. Mit steigender Amplitude wird das Elektron immer mehr lokalisiert, was schliesslich zum Metall-Isolator-Übergang führt. In dieser Arbeit wird die Statistik insbesondere im metallischen Bereich untersucht, und dadurch der Einfluss der Lokalisierung an den Eigenschaften des Systems betrachtet. Zuerst wird die Statistik der Matrixelemente des Dipoloperators untersucht. Die numerischen Ergebnisse für das Anderson-Modell werden mit Vorhersagen der semiklassischen Näherung verglichen. Dann wird der spektrale Strukturfaktor betrachtet, der als Fourier-Transformation der zwei-Punkt Zustandsdichtekorrelationsfunktion definiert wird. Dabei werden besonders die nichtuniversellen Abweichungen von den Vorhersagen der Zufallsmatrixtheorie untersucht. Die Abweichungen werden numerisch ermittelt, und danach mit den analytischen Vorhersagen verglichen. Die Statistik der Wellenfunktionen zeigt ebenfalls Abweichungen von der Zufallsmatrixtheorie. Die Abweichungen sind am größten für Statistik der großen Wellenfunktionsamplituden, die sogenannte seltene Ereignisse darstellen. Die analytischen Vorhersagen für diese Statistik sind teilweise widersprüchlich, und deshalb ist es interessant, sie auch numerisch zu untersuchen.

In large and complex systems, failures can have dramatic consequences, such as black-outs, pandemics or the loss of entire classes of an ecosystem. Nevertheless, it is a centuries-old intuition that by using networks to capture the core of the complexity of such systems, one might understand in which part of a system a phenomenon originates. I investigate this intuition using spectral methods to decouple the dynamics of complex systems near stationary states into independent dynamical modes. In this description, phenomena are tied to a specific part of a system through localized eigenvectors which have large amplitudes only on a few nodes of the system's network. Studying the occurrence of localized eigenvectors, I find that such localization occurs exactly for a few small network structures, and approximately for the dynamical modes associated with the most prominent failures in complex systems. My findings confirm that understanding the functioning of complex systems generally requires to treat them as complex entities, rather than collections of interwoven small parts. Exceptions to this are only few structures carrying exact localization, whose functioning is tied to the meso-scale, between the size of individual elements and the size of the global network. However, while understanding the functioning of a complex system is hampered by the necessary global analysis, the prominent failures, due to their localization, allow an understanding on a manageable local scale. Intriguingly, food webs might exploit this localization of failures to stabilize by causing the break-off of small problematic parts, whereas typical attempts to optimize technological systems for stability lead to delocalization and large-scale failures. Thus, this thesis provides insights into the interplay of complexity and localization, which is paramount to ascertain the functioning of the ever-growing networks on which we humans depend.:1 Introduction 2 Concepts and Tools 2.1 Networks 2.2 Food webs 2.3 Dynamics on networks 2.4 Steady state operating modes 2.5 Bifurcations affecting operating modes 2.6 Dynamical modes 2.7 Generalized models for food webs 3 Perturbation Impact 3.1 Impact of perturbations on food webs 3.2 Examples 3.3 Impact formulation with dynamical modes 3.4 Influence and sensitivity of species 3.5 Localized dynamical modes 3.6 Iterative parameter estimation 3.7 Most important parameters and species 3.8 Discussion 4 Exact Localization 4.1 Graph symmetries 4.2 Localized dynamics on symmetries 4.3 Exactly localized dynamics 4.4 Symmetry reduction in networks 4.5 Application to food webs 4.6 Localization on asymmetric structures 4.7 Nearly-exact localization 4.8 Other systems 4.9 Discussion 5 Approximate Localization 5.1 Spread of a dynamical mode 5.2 Examples for localized instabilities 5.3 Localization of extreme eigenvalues 5.4 Dependence on the system size 5.5 Localization in the model of R. May 5.6 Finding motifs that carry localization 5.7 (Self-)stabilization of food webs 5.8 Repairing localized instabilities 5.9 Discussion 6 Conclusions Acknowledgments Appendix A Parametrization of the Gatun Lake food web B The Master Stability Function approach C Approximate localization on larger structures Bibliography

In der heutigen Zeit spielt die Automatisierung eine immer bedeutendere Rolle, speziell im Bereich der Robotik entwickeln sich immer neue Einsatzgebiete, in denen der Mensch durch autonome Fahrzeuge ersetzt wird. Dabei orientiert sich der Großteil der eingesetzten Roboter an Streckenmarkierungen, die in den Einsatzumgebungen installiert sind. Bei diesen Systemen gibt es jedoch einen hohen Installationsaufwand, was die Entwicklung von Robotersystemen, die sich mithilfe ihrer verbauten Sensorik orientieren, vorantreibt. Es existiert zwar eine Vielzahl an Robotern die dafür verwendet werden können. Die Entwicklung der Steuerungssoftware ist aber immer noch Teil der Forschung. Für die Steuerung wird eine Umgebungskarte benötigt, an der sich der Roboter orientieren kann. Hierfür eignen sich besonders SLAM-Verfahren, die simultanes Lokalisieren und Kartographieren durchführen. Dabei baut der Roboter während seiner Bewegung durch den Raum mithilfe seiner Sensordaten eine Umgebungskarte auf und lokalisiert sich daran, um seine Position auf der Karte exakt zu bestimmen. Im Laufe dieser Arbeit wurden über 30 verschiedene SLAM Implementierungen bzw. Umsetzungen gefunden die das SLAM Problem lösen. Diese sind jedoch größtenteils an spezielle Systembzw. Umgebungsvoraussetzungen angepasste eigenständige Implementierungen. Es existiert keine öffentlich zugängliche Übersicht, die einen Vergleich aller bzw. des Großteils der Verfahren, z.B. in Bezug auf ihre Funktionsweise, Systemvoraussetzungen (Sensorik, Roboterplattform), Umgebungsvoraussetzungen (Indoor, Outdoor, ...), Genauigkeit oder Geschwindigkeit, gibt. Viele dieser SLAMs besitzen Implementierungen und Dokumentationen in denen ihre Einsatzgebiete, Testvoraussetzungen oder Weiterentwicklungen im Vergleich zu anderen SLAMVerfahren beschrieben werden, was aber bei der großen Anzahl an Veröffentlichungen das Finden eines passenden SLAM-Verfahrens nicht erleichtert. Bei einer solchen Menge an SLAM-Verfahren und Implementierungen stellen sich aus softwaretechnologischer Sicht folgende Fragen: 1. Besteht die Möglichkeit einzelne Teile des SLAM wiederzuverwenden? 2. Besteht die Möglichkeit einzelne Teile des SLAM dynamisch auszutauschen? Mit dieser Arbeit wird das Ziel verfolgt, diese beiden Fragen zu beantworten. Hierfür wird zu Beginn eine Übersicht über alle gefundenen SLAMs aufgebaut um diese in ihren grundlegenden Eigenschaften zu unterscheiden. Aus der Vielzahl von Verfahren werden die rasterbasierten Verfahren, welche Laserscanner bzw. Tiefenbildkamera als Sensorik verwenden, als zu untersuchende Menge ausgewählt. Diese Teilmenge an SLAM-Verfahren wird hinsichtlich ihrer nichtfunktionalen Eigenschaften genauer untersucht und versucht in Komponenten zu unterteilen, welche in mehreren verschiedenen Implementierungen wiederverwendet werden können. Anhand der extrahierten Komponenten soll ein Featurebaum aufgebaut werden, der dem Anwender einen Überblick und die Möglichkeit bereitstellt SLAM-Verfahren nach speziellen Kriterien (Systemvoraussetzungen, Umgebungen, ...) zusammenzusetzen bzw. zur Laufzeit anzupassen. Dafür müssen die verfügbaren SLAM Implementierungen und dazugehörigen Dokumentationen in Bezug auf ihre Gemeinsamkeiten und Unterschiede analysiert werden.

In der vorliegenden Publikation wird ein Verfahren beschrieben, dass eine infrastrukturlose Positionierung im Inneren von Gebäuden ermöglicht. Unter infrastrukturlos wird in diesem Zusammenhang die autarke Positionierung eines Systems auf Basis seiner Inertialsensorik ohne den Einsatz von im Gebäude installierter Zusatzlösungen, wie Funksysteme, verstanden. Aufgrund der insbesondere über die Zeit erhöhten Fehlerbehaftung solcher Sensoren werden bei diesem Verfahren Innenraumkarten in den Lokalisierungsprozess einbezogen. Diese Kartendaten erlauben den Ausschluss invalider Positionen und Bewegungen, wie das Durchqueren von Wänden, und ermöglichen somit eine wesentliche Verbesserung der Ortungsgenauigkeit.

Nowadays, satellite-based localization is a well-established technical solution to support several navigation tasks in daily life. Besides the application inside of portable devices, satellite-based positioning is used for in-vehicle navigation systems as well. Moreover, due to its global coverage and the availability of inexpensive receiver hardware it is an appealing technology for numerous applications in the area of Intelligent Transportation Systems (ITSs). However, it has to be admitted that most of the aforementioned examples either rely on modest accuracy requirements or are not sensitive to temporary integrity violations. Although technical concepts of Advanced Driver Assistance Systems (ADASs) based on Global Navigation Satellite Systems (GNSSs) have been successfully demonstrated under open sky conditions, practice reveals that such systems suffer from degraded satellite signal quality when put into urban areas. Thus, the main research objective of this thesis is to provide a reliable vehicle positioning concept which can be used in urban areas without the aforementioned limitations. Therefore, an integrated probabilistic approach which preforms fault detection & exclusion, localization and multi-sensor data fusion within one unified Bayesian framework is proposed. From an algorithmic perspective, the presented concept is based on a probabilistic data association technique with explicit handling of outlier measurements as present in urban areas. By that approach, the accuracy, integrity and availability are improved at the same time, that is, a consistent positioning solution is provided. In addition, a comprehensive and in-depth analysis of typical errors in urban areas within the pseudorange domain is performed. Based on this analysis, probabilistic models are proposed and later on used to facilitate the positioning algorithm. Moreover, the presented concept clearly targets towards mass-market applications based on low-cost receivers and hence aims to replace costly sensors by smart algorithms. The benefits of these theoretical contributions are implemented and demonstrated on the example of a real-time vehicle positioning prototype as used inside of the European research project GAlileo Interactive driviNg (GAIN). This work describes all necessary parts of this system including GNSS signal processing, fault detection and multi-sensor data fusion within one processing chain. Finally, the performance and benefits of the proposed concept are examined and validated both with simulated and comprehensive real-world sensor data from numerous test drives.

As the rapid progress in technology changes our daily life, it also changes how the Industry works. The new developments enable technologies such as the Internet of Moving Things (IoMT), and through these technologies, new challenges arise. IoMT adds one more vital issue, localization, to be solved in comparison to the Internet of Things (IoT). To enable IoMT in the manufacturing industry, there are still problems that need to be overcome. Critical statements such as power consumption, price, accuracy, data management, and size. In this thesis, an evaluation of a new sensor system for an air pneumatic grinder is conducted. The features of the sensor system are to report data from the grinder to the cloud and to localize the position of the grinder. The focus was to optimize the localization algorithm and power consumption of the system. The localization of the grinder was conducted with a new and improved algorithm, Ring Error Difference System (REDS), introduced in this thesis. The new algorithm increased the previous known iRingLA accuracy from 2.91 m to 2.33 m for Bluetooth Low Energy (BLE) and from 3.99 m to 2.84 for Wi-Fi, according to the experiments performed. The final system was able to estimate the operation runtime with an error of 24 s for an operational runtime of 905 s. The operational lifetime of the system was 242 h and 45 h, respectively, for BLE and Wi-Fi. An optimized software was introduced to decrease power consumption. The optimized version was estimated to have an operational lifetime of 1540 h for BLE, which did not reach the wanted lifetime of 3000 h set by Atlas Copco. Hence, I conclude that the hardware, Wemos ESP32, used in the thesis, is not feasible for this solution. Simpler hardware, than the Wemos ESP32, should be used to be able to reach the goal of 3000 h. / De stora framstegen inom dagens teknik förvandlar inte bara vårt dagliga liv det förändrar också tekniken inom industrin. Den nya tekniken möjliggör framsteg så som Internet of Moving Things (IOMT), vilket leder till nya utmaningar. IoMT jämfört med Internet of Things (IoT) lägger till ytterligare utmaningar att lösa så som lokalisering. För att kunna använda IoMT inom tillverkningsindustrin måste ett flertal problem hanteras så som strömförbrukning, pris och noggrannhet på lokaliseringen, datahantering och storlek på systemet. I denna masteruppsatts gör jag en utvärdering av ett nytt sensorsystem för luftdrivna slipmaskiner. Detta sensorsystem rapporterar data från slipmaskinen till molnet och rapporterar positionen av utrustningen. Fokuset på uppsatsen var att optimera lokaliseringsalgoritmen och minska strömförbrukningen för systemet. Lokaliseringen av slipmaskinen gjordes med en ny och förbättrad algoritm, Ring Error Difference System (REDS), som jag introducerar i avhandlingen. Algoritmen förbättrade den tidigare kända RSSI-baserade iRingLA från 2,91 m till 2,33 m med Bluetooth Low Energy (BLE) och från 3,99 m till 2,84 m för Wi-Fi. Det slutliga systemet kunde uppskatta drifttiden med en avvikelse på 24 s av den verkliga drifttiden, 905 s. Systemets operativa livslängd var 242 timmar och 45 timmar för BLE respektive Wi-Fi. Dessutom infördes en optimerad programvara för att minska strömförbrukningen. Den optimerade versionen beräknades ha en livslängd på 1540 timmar för BLE, vilket inte når den önskade livslängden på 3000 timmar satt av Atlas Copco. Ifrån mitt arbete drar jag slutsatsen att hårdvaran som används i uppsatsen, inte kan användas i en slutlig produkt. En enklare hårdvara än Wemos ESP32 bör användas för att kunna nå målet på 3000 timmar.

The dissertation covers phase transitions in the realm of the Anderson model of localization on topologically disordered Voronoi-Delaunay lattices. The disorder is given by random connections which implies correlations due to the restrictive lattice construction. Strictly speaking, the system features "strong anticorrelation", which is responsible for quenched long-range fluctuations of the coordination number. This attribute leads to violations of universal behavior in various system, e.g. Ising and Potts model, and to modifications of the Harris and the Imry-Ma criteria. In general, these exceptions serve to further understanding of critical phenomena. Hence, the question arises whether such deviations also occur in the realm of the Anderson model of localization in combination with random Voronoi-Delaunay lattice. For this purpose, four cases, which are distinguished by the spatial dimension of the systems and by the presence or absence of a magnetic field, are investigated by means of two different methods, i.e the multifractal analysis and the recursive Green function approach. The behavior is classified by the existence and type of occurring phase transitions and by the critical exponent v of the localization length. The results for the four cases can be summarized as follows. In two-dimensional systems, no phase transitions occur without a magnetic field, and all states are localized as a result of topological disorder. The behavior changes under the influence of the magnetic field. There are so-called quantum Hall transitions, which are phase changes between two localized regions. For low magnetic field strengths, the resulting exponent v ≈ 2.6 coincides with established values in literature. For higher strengths, an increased value, v ≈ 2.9, was determined. The deviations are probably caused by so-called Landau level coupling, where electrons scatter between different Landau levels. In contrast, the principle behavior in three-dimensional systems is equal in both cases. Two localization-delocalization transitions occur in each system. For these transitions the exponents v ≈ 1.58 and v ≈ 1.45 were determined for systems in absence and in presence of a magnetic field, respectively. This behavior and the obtained values agree with known results, and thus no deviation from the universal behavior can be observed.:1. Introduction 2. Random Voronoi-Delaunay lattice 2.1. Definition 2.2. Properties 2.3. Numerical construction 3. Anderson localization 3.1. Conventional Anderson transition 3.1.1. Fundamentals 3.1.2. Scaling theory of localization 3.1.3. Universality 3.2. Quantum Hall transition 3.2.1. Universality 3.3. Random Voronoi-Delaunay Hamiltonian 4. Methods 4.1. Multifractal analysis 4.1.1. Fundamentals 4.1.2. Box-size scaling 4.1.3. Partitioning scheme 4.1.4. Numerical realization 4.2. Recursive Green function approach 4.2.1. Fundamentals 4.2.2. Recursive formulation 4.2.3. Layer construction 4.3. Finite-size scaling approach 4.3.1. Scaling functions 4.3.2. Numerical determination 5. Electron behavior on 2D random Voronoi-Delaunay lattices 5.1. 2D orthogonal systems 5.2. 2D unitary systems 5.2.1. Density of states and principal behavior 5.2.2. Criticality in the lowest Landau band 5.2.3. Criticality in higher Landau bands 5.2.4. Edge states 6. Electron behavior on 3D random Voronoi-Delaunay lattices 6.1. 3D orthogonal systems 6.1.1. Pure connectivity disorder 6.1.2. Additional potential disorder 6.2. 3D unitary systems 6.2.1. Pure topological disorder 7. Conclusion Bibliography A. Appendices A.1. Quantum Hall effect on regular lattices A.1.1. Simple square lattice A.1.2. Triangular lattice A.2. Further quantum Hall transitions on 2D random Voronoi-Delaunay lattices Lebenslauf Publications / Diese Dissertation behandelt Phasenübergange im Rahmen des Anderson-Modells der Lokalisierung in topologisch ungeordneten Voronoi-Delaunay-Gittern. Die spezielle Art der Unordnung spiegelt sich u.a. in zufälligen Verknüpfungen wider, welche aufgrund der restriktiven Gitterkonstruktion miteinander korrelieren. Genauer gesagt zeigt das System eine "starke Antikorrelation", die dafür sorgt, dass langreichweitige Fluktuationen der Verknüpfungszahl unterdrückt werden. Diese Eigenschaft hat in anderen Systemen, z.B. im Ising- und Potts-Modell, zur Abweichung vom universellen Verhalten von Phasenübergängen geführt und bewirkt eine Modifikation von allgemeinen Aussagen, wie dem Harris- and Imry-Ma-Kriterium. Die Untersuchung solcher Ausnahmen dient zur Weiterentwicklung des Verständnisses von kritischen Phänomenen. Somit stellt sich die Frage, ob solche Abweichungen auch im Anderson-Modell der Lokalisierung unter Verwendung eines solchen Gitters auftreten. Dafür werden insgesamt vier Fälle, welche durch die Dimension des Gitters und durch die An- bzw. Abwesenheit eines magnetischen Feldes unterschieden werden, mit Hilfe zweier unterschiedlicher Methoden, d.h. der Multifraktalanalyse und der rekursiven Greensfunktionsmethode, untersucht. Das Verhalten wird anhand der Existenz und Art der Phasenübergänge und anhand des kritischen Exponenten v der Lokalisierungslänge unterschieden. Für die vier Fälle lassen sich die Ergebnisse wie folgt zusammenfassen. In zweidimensionalen Systemen treten ohne Magnetfeld keine Phasenübergänge auf und alle Zustände sind infolge der topologischen Unordnung lokalisiert. Unter Einfluss des Magnetfeldes ändert sich das Verhalten. Es kommt zur Ausformung von Landau-Bändern mit sogenannten Quanten-Hall-Übergängen, bei denen ein Phasenwechsel zwischen zwei lokalisierten Bereichen auftritt. Für geringe Magnetfeldstärken stimmen die erzielten Ergebnisse mit den bekannten Exponenten v ≈ 2.6 überein. Allerdings wurde für stärkere magnetische Felder ein höherer Wert, v ≈ 2.9, ermittelt. Die Abweichungen gehen vermutlich auf die zugleich gestiegene Unordnungsstärke zurück, welche dafür sorgt, dass Elektronen zwischen verschiedenen Landau-Bändern streuen können und so nicht das kritische Verhalten eines reinen Quanten-Hall-Überganges repräsentieren. Im Gegensatz dazu ist das Verhalten in dreidimensionalen Systemen für beide Fälle ähnlich. Es treten in jedem System zwei Phasenübergänge zwischen lokalisierten und delokalisierten Bereichen auf. Für diese Übergänge wurde der Exponent v ≈ 1.58 ohne und v ≈ 1.45 unter Einfluss eines magnetischen Feldes ermittelt. Dieses Verhalten und die jeweils ermittelten Werte stimmen mit bekannten Ergebnissen überein. Eine Abweichung vom universellen Verhalten wird somit nicht beobachtet.:1. Introduction 2. Random Voronoi-Delaunay lattice 2.1. Definition 2.2. Properties 2.3. Numerical construction 3. Anderson localization 3.1. Conventional Anderson transition 3.1.1. Fundamentals 3.1.2. Scaling theory of localization 3.1.3. Universality 3.2. Quantum Hall transition 3.2.1. Universality 3.3. Random Voronoi-Delaunay Hamiltonian 4. Methods 4.1. Multifractal analysis 4.1.1. Fundamentals 4.1.2. Box-size scaling 4.1.3. Partitioning scheme 4.1.4. Numerical realization 4.2. Recursive Green function approach 4.2.1. Fundamentals 4.2.2. Recursive formulation 4.2.3. Layer construction 4.3. Finite-size scaling approach 4.3.1. Scaling functions 4.3.2. Numerical determination 5. Electron behavior on 2D random Voronoi-Delaunay lattices 5.1. 2D orthogonal systems 5.2. 2D unitary systems 5.2.1. Density of states and principal behavior 5.2.2. Criticality in the lowest Landau band 5.2.3. Criticality in higher Landau bands 5.2.4. Edge states 6. Electron behavior on 3D random Voronoi-Delaunay lattices 6.1. 3D orthogonal systems 6.1.1. Pure connectivity disorder 6.1.2. Additional potential disorder 6.2. 3D unitary systems 6.2.1. Pure topological disorder 7. Conclusion Bibliography A. Appendices A.1. Quantum Hall effect on regular lattices A.1.1. Simple square lattice A.1.2. Triangular lattice A.2. Further quantum Hall transitions on 2D random Voronoi-Delaunay lattices Lebenslauf Publications

Visual place recognition is the task of finding same places in a set of database images for a given set of query images. This becomes particularly challenging for long-term applications when the environmental condition changes between or within the database and query set, e.g., from day to night. Visual place recognition in changing environments can be used if global position data like GPS is not available or very inaccurate, or for redundancy. It is required for tasks like loop closure detection in SLAM, candidate selection for global localization, or multi-robot/multi-session mapping and map merging. In contrast to pure image retrieval, visual place recognition can often build upon additional information and data for improvements in performance, runtime, or memory usage. This includes additional data-inherent knowledge about information that is contained in the image sets themselves because of the way they were recorded. Using data-inherent knowledge avoids the dependency on other sensors, which increases the generality of methods for an integration into many existing place recognition pipelines. This thesis focuses on the usage of additional data-inherent knowledge. After the discussion of basics about visual place recognition, the thesis gives a systematic overview of existing data-inherent knowledge and corresponding methods. Subsequently, the thesis concentrates on a deeper consideration and exploitation of four different types of additional data-inherent knowledge. This includes 1) sequences, i.e., the database and query set are recorded as spatio-temporal sequences so that consecutive images are also adjacent in the world, 2) knowledge of whether the environmental conditions within the database and query set are constant or continuously changing, 3) intra-database similarities between the database images, and 4) intra-query similarities between the query images. Except for sequences, all types have received only little attention in the literature so far. For the exploitation of knowledge about constant conditions within the database and query set (e.g., database: summer, query: winter), the thesis evaluates different descriptor standardization techniques. For the alternative scenario of continuous condition changes (e.g., database: sunny to rainy, query: sunny to cloudy), the thesis first investigates the qualitative and quantitative impact on the performance of image descriptors. It then proposes and evaluates four unsupervised learning methods, including our novel clustering-based descriptor standardization method K-STD and three PCA-based methods from the literature. To address the high computational effort of descriptor comparisons during place recognition, our novel method EPR for efficient place recognition is proposed. Given a query descriptor, EPR uses sequence information and intra-database similarities to identify nearly all matching descriptors in the database. For a structured combination of several sources of additional knowledge in a single graph, the thesis presents our novel graphical framework for place recognition. After the minimization of the graph's error with our proposed ICM-based optimization, the place recognition performance can be significantly improved. For an extensive experimental evaluation of all methods in this thesis and beyond, a benchmark for visual place recognition in changing environments is presented, which is composed of six datasets with thirty sequence combinations.

Movement of motor proteins along cytoskeletal filaments is fundamental for various cellular processes ranging from muscle contraction over cell division and flagellar movement to intracellular transport. Not surprisingly, the impairment of motility was shown to cause severe diseases. For example, a link between impaired intracellular transport and neurodegenerative diseases, such as Alzheimer’s, has been established. There, the movement of kinesin-1, a neuronal motor protein transporting vesicles along microtubules toward the axonal terminal, is thought to be strongly affected by roadblocks leading to malfunction and death of the nerve cell. Detailed information on how the motility of kinesin-1 deteriorates in the presence of roadblocks and whether the motor has a mechanism to circumvent such obstructions is scarce. In this thesis, kinesin-1 motility was studied in vitro in the presence of rigor kinesin-1 mutants, which served as permanent roadblocks, under controlled single-molecule conditions. The 25 nm wide microtubule track, consisting of 13 individual protofilaments, resembles a multi-lane environment for transport by processive kinesin-1 motors. The existence of multiple traffic-lanes, allows kinesin-1 to utilize different paths for cargo transport and potentially also for the circumvention of roadblocks. However, direct observation of motor encounters with roadblocks has been intricate in the past, mainly due to limitations in both, spatial and temporal resolution. These limitations, intrinsic to fluorescent probes commonly utilized to report on the motor positions, originate from a low rate of photon generation (low brightness) and a limited photostability (short observation time). Thus, studying kinesin-1 encounters with microtubule-associated roadblocks requires alternative labels, which explicitly avoid the shortcomings of fluorescence and consequently allow for a higher localization precision. Promising candidates for replacing fluorescent dyes are gold nanoparticles (AuNPs), which offer an enormous scattering cross-section due to plasmon resonance in the visible part of the optical spectrum. Problematic, however, is their incorporation into conventionally used (fluorescence) microscopes, because illumination and scattered light have the same wavelength and cannot be separated spectrally. Therefore, an approach based on total internal reflection (TIR) utilizing a novel parabolically shaped quartz prism for illumination was developed within this thesis. This approach provided homogenous and spatially invariant illumination profiles in combination with a convenient control over a wide range of illumination angles. Moreover, single-molecule fluorescence as well as single-particle scattering were detectable with high signal-to-noise ratios. Importantly, AuNPs with a diameter of 40 nm provided sub-nanometer localization accuracies within millisecond integration times and reliably reported on the characteristic 8 nm stepping of individual kinesin-1 motors moving along microtubules. These results highlight the potential of AuNPs to replace fluorescent probes in future single-molecule experiments. The newly developed parabolic prism-type TIR microscope is expected to strongly facilitate such approaches in the future. To study how the motility of kinesin-1 is affected by permanent roadblocks on the microtubule lattice, first, conventional objective-type TIRF microscopy was applied to GFP-labeled motors. An increasing density of roadblocks caused the mean velocity, run length, and dwell time to decrease exponentially. This is explained by (i) the kinesin-1 motors showing extended pausing phases when confronted with a roadblock and (ii) the roadblocks causing a reduction in the free path of the motors. Furthermore, kinesin-1 was found to be highly sensitive to the crowdedness of microtubules as a roadblock decoration as low as 1 % sufficed to significantly reduce the landing rate. To study events, where kinesin-1 molecules continued their runs after having paused in front of a roadblock, AuNPs were loaded onto the tails of the motors. When observing the kinesin-1 motors with nanometer-precision, it was interestingly found that about 60 % of the runs continued by movements to the side, with the left and right direction being equally likely. This finding suggests that kinesin-1 is able to reach to a neighboring protofilament in order to ensure ongoing transportation. In the absence of roadblocks, individual kinesin-1 motors stepped sideward with a much lower, but non-vanishing probability (0.2 % per step). These findings suggest that processive motor proteins may possess an intrinsic side stepping mechanism, potentially optimized by evolution for their specific intracellular tasks.

Superpixels are the results of an image oversegmentation. They are an established intermediate level image representation and used for various applications including object detection, 3d reconstruction and semantic segmentation. While there are various approaches to create such segmentations, there is a lack of knowledge about their properties. In particular, there are contradicting results published in the literature. This thesis identifies segmentation quality, stability, compactness and runtime to be important properties of superpixel segmentation algorithms. While for some of these properties there are established evaluation methodologies available, this is not the case for segmentation stability and compactness. Therefore, this thesis presents two novel metrics for their evaluation based on ground truth optical flow. These two metrics are used together with other novel and existing measures to create a standardized benchmark for superpixel algorithms. This benchmark is used for extensive comparison of available algorithms. The evaluation results motivate two novel segmentation algorithms that better balance trade-offs of existing algorithms: The proposed Preemptive SLIC algorithm incorporates a local preemption criterion in the established SLIC algorithm and saves about 80 % of the runtime. The proposed Compact Watershed algorithm combines Seeded Watershed segmentation with compactness constraints to create regularly shaped, compact superpixels at the even higher speed of the plain watershed transformation. Operating autonomous systems over the course of days, weeks or months, based on visual navigation, requires repeated recognition of places despite severe appearance changes as they are for example induced by illumination changes, day-night cycles, changing weather or seasons - a severe problem for existing methods. Therefore, the second part of this thesis presents two novel approaches that incorporate superpixel segmentations in place recognition in changing environments. The first novel approach is the learning of systematic appearance changes. Instead of matching images between, for example, summer and winter directly, an additional prediction step is proposed. Based on superpixel vocabularies, a predicted image is generated that shows, how the summer scene could look like in winter or vice versa. The presented results show that, if certain assumptions on the appearance changes and the available training data are met, existing holistic place recognition approaches can benefit from this additional prediction step. Holistic approaches to place recognition are known to fail in presence of viewpoint changes. Therefore, this thesis presents a new place recognition system based on local landmarks and Star-Hough. Star-Hough is a novel approach to incorporate the spatial arrangement of local image features in the computation of image similarities. It is based on star graph models and Hough voting and particularly suited for local features with low spatial precision and high outlier rates as they are expected in the presence of appearance changes. The novel landmarks are a combination of local region detectors and descriptors based on convolutional neural networks. This thesis presents and evaluates several new approaches to incorporate superpixel segmentations in local region detection. While the proposed system can be used with different types of local regions, in particular the combination with regions obtained from the novel multiscale superpixel grid shows to perform superior to the state of the art methods - a promising basis for practical applications.