Demonstratoren sind äußerst hilfreiche Werkzeuge, welche der Veranschaulichung neuer Technologien dienen. Besonders im Kontext von Living Labs, in denen es um Bildung, Forschung, gemeinsame Entwicklung und Experimente geht, können sie von großer Bedeutung sein. Sie unterliegen in ihrer Umsetzung besonderen Anforderungen, die sich stark von denen an Produkte für den Markt unterscheiden. Diese Arbeit stellt das Konzept und die Umsetzung eines Smart Mirror Demonstrators vor, der für das Living Lab des KI-Kompetenzzentrums ScaDS.AI in Leipzig entwickelt wurde. Der Demonstrator veranschaulicht einen Anwendungsfall von Objekt- und Ähnlichkeitserkennung, die auf künstlichen neuronalen Netzen basieren. Er besteht aus einem Smart Mirror, der Kleidungsstücke erkennen und visuell ähnliche Kleidungsstücke vorschlagen kann. Ein Nutzer kann mit dem Demonstrator ein Foto von sich aufnehmen, auf dem die erkannten Kleidungsstücke mit sogenannten Bounding Boxes markiert werden. Anschließend werden visuell ähnliche Kleidungsstücke aus einer Datenbank herausgefiltert und angezeigt. In der Arbeit geht es zum einen um die verwendeten Technologien, als auch um die Architektur des Demonstrators selbst. Die Softwarearchitektur des Demonstrators orientiert sich an dem Aufbau von Webanwendungen und besteht aus einem Backend und einem Frontend. Das Backend wurde als REST-API umgesetzt, die in Python unter Verwendung des Frameworks Flask geschrieben wurde. Das Frontend besteht aus einer Single Page Application, umgesetzt mit dem Javascript Framework Vue. Beide Anwendungen wurden mit Hilfe der Software Docker in Containern umgesetzt, um die Laufzeitumgebung zu virtualisieren und die Ausführung auf verschiedenen Rechnern zu erleichtern. Der Technologie-Stack bestehend aus Flask, Vue und Docker und die Umsetzung unter Beachtung der REST Prinzipien sind positiv zu bewerten und wurden im ScaDS.AI in Anlehnung an den be- schriebenen Demonstrator zum Standard für die Entwicklung weiterer Demonstratoren.:1. Einleitung 1.1. Projektrahmen 1.2. Hintergrund und Idee 1.3. Aufbau der Arbeit 2. Anforderungen 2.1. Funktionale Anforderungen 2.1.1. Objekterkennung 2.1.2. Kleidungsvorschläge 2.1.3. Visualisierung und Benutzung 2.1.4. Hardware 2.2. Projektrahmen und nicht-funktionale Anforderungen 2.2.1. Das ScaDS.AI und Living Lab in Leipzig 2.2.2. Demonstratoren 2.3. Zusammenfassung 3. Technische Grundlagen 3.1. Objekterkennung 3.1.1. Convolutional Neural Networks 3.1.2. Single Shot Multibox Detector 3.2. Architektur und Technologie-Stack 3.2.1. REST Paradigma 3.2.2. Python Webframework Flask 3.2.3. Single Page Applications mit Vue 3.2.4. Virtualisierung mit Docker 3.3. Zusammenfassung 4. Implementierung 4.1. Architekturentwurf 4.2. Backend Programmierung 4.2.1. Umsetzung als Flask App 4.2.2. Detection 4.2.3. Matching 4.2.4. Umsetzung der REST API 4.3. Frontend Programmierung 4.3.1. Darstellungsschicht 4.3.2. Serviceschicht 4.3.3. Speicherschicht 4.3.4. Datenfluss 4.3.5. Online-Version 4.4. Containerisierung 4.5. Hardware 4.5.1. Samsung Flip 4.5.2. Kamera 4.5.3. Rechner 4.6. Zusammenfassung 5. Projektauswertung 5.1. Erfüllung der Anforderungen 5.2. Schwierigkeiten 5.3. Zusammenfassung 6. Ausblick 7. Literaturverzeichnis 8. Abbildungsverzeichnis 9. Quelltextverzeichnis 10. Anhang

Most domestication hypotheses propose that dogs have been selected for enhanced communication and interactions with humans, including learning socially from human demonstrators. However, to what extent these skills are newly derived and to what extent they originate from wolf–wolf interactions is unclear. In order to test for the possible origins of dog social cognition, we need to compare the interactions of wolves and dogs with humans and with conspecifics. Here, we tested identically raised and kept juvenile wolves and dogs in a social learning task with human and conspecific demonstrators. Using a local enhancement task, we found that both wolves and dogs benefitted from a demonstration independent of the demonstrator species in comparison to a control, no demonstration condition. Interestingly, if the demonstrator only pretended to hide food at the target location, wolves and dogs reacted differently: while dogs differentiated between this without-food and with-food demonstration independent of the demonstrator species, wolves only did so in case of human demonstrators. We attribute this finding to wolves being more attentive toward behavioral details of the conspecific models than the dogs: although the demonstrator dogs were trained to execute the demonstration, they disliked the food reward, which might have decreased the interest of the wolves in finding the food reward. Overall, these results suggest that dogs but also wolves can use information provided by both human and conspecific demonstrators in a local enhancement task. Therefore we suggest that a more fine-scale analysis of dog and wolf social learning is needed to determine the effects of domestication.

An observer rat that interacts with a conspecific, a demonstrator that has eaten a flavoured food, is subsequently more likely to eat that food than an alternative, novel food (Galef & Wigmore, 1983). In the first part of this thesis, four experiments were undertaken to determine the influence of unreliable demonstrators on observer food preference. In the first three experiments, observers were poisoned after interaction with demonstrators; a demonstrator that demonstrated a food that led to poisoning was considered an unreliable demonstrator. The first experiment compared the influence of an unreliable demonstrator and an unfamiliar demonstrator on observers' food preferences. Experiment 2 compared the influence of demonstration from an unreliable conspecific and a familiar conspecific that had only provided irrelevant information. The third experiment compared the influence of an unreliable and a reliable demonstrator. Experiment 4 compared the influence for protein-deficient observers of a demonstrator that had previously demonstrated a protein-deficient diet and a demonstrator that was unfamiliar. None of the first four experiments showed a significant difference in effectiveness of social transmission due to demonstrator quality. In the second part of the thesis I investigated the influence of familiarity on demonstrator effectiveness. In Experiment 5, the influence of local sisters was compared with that of unfamiliar non-relatives as demonstrators. Sisters were not better demonstrators than unfamiliar non-relatives. / Thesis / Master of Science (MSc)

Ein Carbonbeton-Schalentragwerk, das im Rahmen des DFG-Schwerpunktprogramms SPP 1542 „Leicht Bauen mit Beton“ von der TU Chemnitz entwickelt wurde, wurde nach Projektabschluss erfolgreich errichtet. / A carbon reinforced concrete shell structure which wasdeveloped by Chemnitz University of Technology as part of the DFG Priority Programme SPP 1542 “Concrete ligth” was successfully erected after the project was completed.

La recherche de configurations d’aéronefs plus efficaces mène les ingénieurs à explorer de nouveaux concepts tels que l’aile volante, l’aile haubanée ou l’aile en jointive. Contrairement à la configuration classique aile-fuselage, qui est bien connue et étudiée, le comportement en vol de ces nouveaux concepts d'avion est peu connu. Dans ce contexte, la conception, la construction et les essais de modèles à l'échelle aéroélastiquement semblables se présentent comme un moyen peu risqué d'acquérir des connaissances expérimentales sur ces nouveaux concepts. Un modèle aéroélastiquement semblable présente le même comportement aéroélastique (mis à l’échelle) que l’avion de référence à échelle réelle. En général, le même comportement aéroélastique implique de reproduire les mêmes déplacements pour des conditions du flux d’air données, ainsi que les mêmes vitesses de flottement ou de divergence statique mises à l'échelle. Pour résoudre le problème de similitude, l'approche est divisée en trois parties. Dans le premier cas, nous traitons le problème de similitude aéroélastique lorsque les paramètres de similitude du flux aérodynamique peuvent être complètement préservés. Dans cette situation, le problème consiste simplement à reproduire la réponse dynamique modale de l’aile mise à l'échelle en optimisant les propriétés de la structure et de la masse. Dans la deuxième partie, nous nous concentrons sur l’optimisation du design de la forme de l’aile pour reproduire la réponse du flottement lorsque les paramètres de remise à l’échelle du flux aérodynamique ne peuvent pas être atteints. / The search for more efficient aircraft configurations leads designers to explore new concepts such as the blended wing body, the strut-braced wing, or the box wing. Unlike the classical wing-fuselage configuration, which is well known and understood, few is known about the in-flight behavior of these new aircraft concepts. In that context, the design, construction, and testing of unmanned aeroelastically scaled models presents itself as a low-risk means of acquiring experimental knowledge on these new concepts. An aeroelastically scaled model exhibits the same scaled aeroelastic behavior as the full-scale reference aircraft. Typically, the same aeroelastic behavior implies matching the displacements for some given scaled airflow conditions, as well as the scaled flutter or static divergence speeds. To address the similarity problem, we divide the approach in three parts. In the first one we deal with the aeroelastic similarity problem when the aerodynamic flow scaling conditions can be completely preserved. In that situation, the problem is reduced to simply matching the scaled modal dynamic response of the wing through optimization of the structure and mass properties. In the second part, we focus on the wing planform design optimization to match the flutter response when the airflow scaling parameters cannot be achieved.

This master thesis was performed at the business area of Aeronautics at Saab AB in collaboration with Linköping University during the spring of 2015. In a complex product development environment, having knowledge about different processes is advantageous for efficiency. Model Based Definition (MBD) is a product development process where a 3D-model is the main source of information and the same 3D-model is applied all the way from design to production. In assembly, the operator follows work instructions where the 3D-model, and its requirements, is visualized. The model is always updated to the latest version and no 2D-drawings are needed. Saab applied MBD during the development of the new generation of the fighter aircraft JAS 39 Gripen. This change, from previously 2D to 3D, has caused that both internal and external people have minor knowledge about the new developing process of MBD. The purpose of the thesis was to develop a demonstrator acting as an educational environment to share knowledge about the MBD-process and the 3D work instructions. New methods and processes could be tested and evaluated in the demonstrator before implementing into the real product development process. By following and developing an interdependent and iterative product development process, this work has visualized the MBD-process. Designing and developing a demonstrator, using the same tools as in the real product development process, accomplished this. This thesis has developed a demonstrator that includes the main components of a physical model with corresponding 3D work instructions and a conceptual layout. A physical Lego model of Gripen provides a flexible and interesting way of sharing knowledge to the user who interacts with the demonstrator. The 3D work instructions were created in a way so that the user can assemble and interact with the same expressions and terms in order to get an understanding about how they are used. The educational aspect is important where simplifications and additional notes to the instructions help to get a better understanding. Depending on who the user is, different levels of preparations are needed. The recognition factor is important to a user with experience of MBD; it needs to be able to understand how different terms and requirements are used in the development process. The future work is about setting up the demonstrator and conduct usability tests to evaluate, modify and implement more details. Using a demonstrator in this purpose can be helpful for evaluating different techniques, methods or systems and reduce the errors in the product development process. It can also encourage people to a new enjoyable way of learning.

Eurofins is one of the world's largest laboratories which, among other things, offer chemical and microbiological analyses in agriculture, food and environment. Several 100.000 tests of various foods are executed each year at Eurofins’ facility in Jönköping and the current processes include much repeated manual tasks which could cause ergonomic problems. The company therefore wants to investigate the possibilities of utilizing Human-Robot Collaboration (HRC) at their facility. Human-Robot Collaboration is a growing concept that has made a big impression in both robot development and Industry 4.0. A HRC approach allow humans and robots to share their workspaces and work side by side, without being separated by a protective fence which is common among traditional industrial robots. Human-Robot Collaboration is therefore believed to be able to optimize the workflows and relieve human workers from unergonomic tasks. The overall aim of the research project presented is to help the company to gain a better understanding about the existing HRC technologies. To achieve this goal, the state-of-the-art of HRC had to be investigated and the needs, possibilities and limitations of HRC applications had to be identified at Eurofins’ facility. Once these have been addressed, a demonstrator could be built which could be used for evaluating the applicability and suitability of HRC at Eurofins. The research project presented used the design science research process. The state-of-the-art of HRC was studied in a comprehensive literature review, reviewing sterile robots and mobile robotics as well. The presented literature review could identify possible research gaps in both HRC in laboratory environments and mobile solutions for HRC applications. These areas studied in the literature review formed together the basis of the prepared observations and interviews, used to generate the necessary data to develop the design science research artefact, the demonstrator. ABB's software for robotic simulation and offline programming, RobotStudio, were used in the development of the demonstrator, with the collaborative robot YuMi chosen for the HRC implementation. The demonstrator presented in the research project has been built, tested and refined in accordance to the design science research process. When the demonstrator could illustrate an applicable solution, it was evaluated for its performance and quality using a mixed methods approach. Limitations were identified in both the performance and quality of the demonstrator's illustrated HRC implementation, including adaptability and sterility constraints. The research project presented could conclude that a HRC application would be possible at a station which were of interest by the company, but would however not be recommended due to the identified constraints. Instead, the company were recommended to look for stations which are more standardized and have less hygienic requirements. By the end of the research project, additional knowledge was contributed to the company, including how HRC can affect today's working methods at Eurofins and in laboratory environments in general.

Resource allocation consists in allocating spectrum and power on every link of a network, possibly under power and rate requirements. In the context of cognitive radios, almost 15 years of research produced an impressive amount of theoretical contributions, exploring a wide range of possibilities. However, despite the ever-growing list of imaginable scenarios, we observe in Chapter 2 that most of these studies are based on similar working hypotheses. Our first contribution is to challenge some of these hypotheses, and propose a novel resource allocation scheme. Sticking to realistic assumptions, we show how our scheme reduces both computational complexity and control traffic, compared to other state-of-the-art techniques.<p><p>Due to a majority of the abovementioned studies making some constraining assumptions, realistic system designs and experimental demonstrations are much more quiet and unharvested fields. In an effort to help this transition from theory to practice, our second contribution is a four-nodes cognitive network demonstrator, presented in Chapter 3. In particular, we aim at providing a modular platform available for further open collaboration: different options for spectrum sensing, resource allocation, synchronisation and others can be experimented on this demonstrator. As an example, we develop a simple protocol to show that our proposed resource allocation scheme is fully implementable, and that primary users can be avoided using our approach.<p><p>Chapter 4 aims at removing another working hypothesis made when developping our resource allocation scheme. Indeed, resource alloca- tion is traditionally a Media Access Control (MAC) layer problem. This means that when solving a resource allocation problem in a network, the routing paths are usually assumed to be known. Conversely, the routing problem, which is a network layer issue, usually assumes that the available capacities on each link of the network (which depend on resource allocation) are known. Nevertheless, these two problems are mathematically entangled, and a cross-layer allocation strategy can best decoupled approaches in several ways, as we discuss in Chapter 4. Accordingly, our third and last contribution is to develop such a cross-layer allocation scheme for the scenario proposed in previous chapters.<p><p>All conclusions are summarised in Chapter 5, which also points to a few tracks for future research. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

Social observational learning is one of learning abilities expected in domestic horses (Equus caballus) because of their ecological and evolutional history. However, a few studies focused on this type of learning in horses failed to provide clear evidence of observational learning and/or could not distinguished it from other types of learning. We tested interspecific observational learning abilities using the spatial task and a human demonstrator. We hypothesised that 1) horses with possibility of observing a human demonstrator will complete the task in shorter time than control horses without any demonstrator, and 2) horses observing a familiar demonstrator will carry out the task in shorter time than horses with an unfamiliar demonstrator due to established positive human - horse relationship. Twenty - four riding horses of mixed age and breed were randomly allocated to three groups per 8 and started the task either with observing a familiar demonstrator, unfamiliar demonstrator or no demonstrator (control group). Each horse was released individually at the starting point in the experimental paddock and the latency to pass the task was recorded. A horse completed the task once it walked 25 m from the starting point to the squared area (4x4 m) fenced by a tape, went into it through the entrance on the...

This project seeks to demonstrate routing optimization that would allow pedestrian and micromobility user groups to select and prioritize different route features according to their preferences. Through the creation of a routing demonstrator that considers both static and dynamic parameters in the form of pavement quality, elevation climb, travel time, and air quality, along with user-specified weights for their prioritization of each of these parameters, a number of routes were created and mapped to qualitatively compare against routes representing only a shortest path. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>