The Diploma thesis describes the design, manufacture and testing of a universal automation machine for church bells. The research section summarizes the bell, way of use and properties. The drives and their starting and power switching characteristic are analyzed. The chain drive based on the selected engine and the whole assembly is designed and tested by using PLC. Unique controller based on a modular design was designed. Sensors based on magnetic rotary encoder were also created for this unit. The entire circuit was tested on three evangelical church bells in the town Klobouky u Brna.

The goal of this diploma thesis is a proposal of autonomous two wheeled balancing robot, differentially driven. This kind of robot is especially suitable in confined space, where it can utilize its maneuver skills. Many criteria as operational conditions, materials, size and weight of the robot, suitable hardware and sensors must to be considered, when designing the robot. Development and implementation of autonomous balancing control system is also part of the thesis.

This thesis deals with the design of the construction of a pendulum of a single toggle jaw crusher. The first part of this thesis treats the mechanical raw minerals disintegration. In the second part the characteristics of a single toggle jaw crusher and double toggle jaw crusher are compared. The third part addresses the calculation and the construction of the main components of the jaw crusher. Next part of the thesis consists of the calculation of the crushing force, the pendulum design, and the method of manufacturing a pendulum. The strength analysis of the designed pendulum using the finite element method (FEM) in program NX-Ideas is made in the last part of this thesis.

This report presents the construction of a two-wheeled self balancing robot with the ability to handle an uneven load. Two-wheeled self-balancing robots have proven to be a potential solution to the problem of efficient warehouse management, thanks to their ability to navigate tight spaces quickly and efficiently while balancing their load. The research questions defined include the ability to develop an algorithm to dynamically adjust the robot’s tilt angle, how an uneven load affects the robot’s stability, and whether a PID controller is sufficient in this context. The project was limited to constructing a prototype that could balance with an external load, with a budget of 1,000 Swedish kronor and a four-month timeframe. By testing and evaluating different control algorithms, the robot’s performance was presented in terms of stability and efficiency. The implementation of PID control was successful, and the robot was able to balance as a result. However, the goal of handling an uneven load was not met without the implementation of an additional algorithm to dynamically adjust the robot’s tilt angle. With these two control techniques, the robot was able to balance with and without an added load with good stability. To evaluate performance, tests were performed with the load placed centered and off-centered on the robot’s top plate. The results of the tests showed that the robot was able to dynamically adjust its tilt angle to balance with added weight without affecting stability. / Denna rapport presenterar konstruktionen av en tvåhjulig självbalanserande robot med förmågan att hantera snedfördelad last. Tvåhjuliga självbalanserande robotar har visat sig vara en potentiell lösning på problemet kring effektiv lagerhantering tack vare deras förmåga att hantera snäva utrymmen på ett snabbt och energieffektivt sätt samtidigt som lasten balanseras. Forskningsfrågorna som definierades inkluderar möjligheten att framställa en algoritm för att dynamiskt ställa in robotens lutningsvinkel, hur ojämn last påverkar robotens stabilitet, och om en PID-kontroller är tillräcklig i detta sammanhang. Projektet begränsades till att konstruera en prototyp som klarar av att balansera med extern last, med en budget på 1000 svenska kronor och en tidsram på fyra månader. Genom att testa och utvärdera olika kontrollalgoritmer presenterades robotens prestanda i termer av stabilitet och effektivitet. Roboten balanserade tack vare implementeringen av PID-reglering, men önskemålet om snedfördelad last uppfylldes inte och det krävdes ytterligare en algoritm för att dynamiskt reglera robotens lutningsvinkel. Med hjälp av dessa två reglertekniker kunde roboten balansera både med och utan adderad last med god stabilitet. Tester utfördes för att utvärdera prestandan när lasten var placerad både centrerat och ocentrerat på robotens topplatta. Resultaten visade att roboten kan dynamiskt anpassa lutningsvinkeln för att balansera med tillagd vikt utan att stabiliteten påverkas.

This dissertation presents a self-contained study of periodically-driven quantum systems. Following a brief introduction to Floquet theory, we introduce the inverse-frequency expansion, variants of which include the Floquet-Magnus, van Vleck, and Brillouin-Wigner expansions. We reveal that the convergence properties of these expansions depend strongly on the rotating frame chosen, and relate the former to the existence of Floquet resonances in the quasienergy spectrum. The theoretical design and experimental realisation (`engineering') of novel Floquet Hamiltonians is discussed introducing three universal high-frequency limits for systems comprising single-particle and many-body linear and nonlinear models. The celebrated Schrieffer-Wolff transformation for strongly-correlated quantum systems is generalised to periodically-driven systems, and a systematic approach to calculate higher-order corrections to the Rotating Wave Approximation is presented. Next, we develop Floquet adiabatic perturbation theory from first principles, and discuss extensively the adiabatic state preparation and the corresponding leading-order non-adiabatic corrections. Special emphasis is thereby put on geometrical and topological objects, such as the Floquet Berry curvature and the Floquet Chern number obtained within linear response in the presence of the drive. Last, pre-thermalisation and thermalisation in closed, clean periodically-driven quantum systems are studied in detail, with the focus put on the crucial role of Floquet many-body resonances for energy absorption.

This report is the result of Master of Science thesis project developed for KTH Royal Institute of Technology in collaboration with the Forestry Research Institute of Sweden (Skogforsk) for the Forestry Master Thesis School 2016. In the past few years, Adams MBD models have been created to verify the better comfortability of six-wheeled pendulum-arm-suspension compared to bogie-suspended forwarder. Based on the current achievements, this thesis would focus on discovering the energy consumption states of both types of forwarder. A study on the energy usage of a pendulum arm suspension forwarder was preformed based on existing design parameters and test data gathered by Skogforsk, providing insight about the performance of XT28 when operating on a test track. Hydraulic and mechanical models of the forwarder were built using Adams and Simulink/Matlab. These models were used for the simulation of the working process and calculation of the energy consumption. The result of this research project shows that active pendulum arm suspension forwarder saves approximately 28% of energy consumption compared to passive suspension. It is also shown that a speed increase from 0.84m/s to 1m/s increases the energy consumptio with approximately 10%. / Rapporten är resultatet av ett masterprojekt som utförts vid KTH Kungliga Tekniska Högskolan i samarbete med Skogforsk för Skogsforskningskolan 2017. Under de senaste åren har ett antal Adams-baserade MBD-modeller skapats, med syftet att undersöka och verifiera hur mycket vibrationsmiljön kan förbättras med en sexhjulig pendelarmsdämpad skotare jämfört med en traditionell bogie-maskin. Denna avhandling fokuserade på att jämföra energiförbrukningen hos båda typer av maskiner. En studie om energiförbrukningen hos en pendularmsskotare har utförts, baserat på befintliga designparametrar och provningsdata insamlade av Skogforsk, vilket gav insikt om energiprestandan hos den aktivt dämpade pendelarmsskotaren XT28 vid drift på provbana. Hydrauliska och mekaniska modeller av maskinen skapades med Adams och Matlab/Simulink. Dessa modeller användes för simulering av en idealiserad arbetsprocess och beräkning av energiförbrukningen. Resultatet av detta projekt visar att aktivt dämpad pendularmsskotare har cirka 28% lägre energiförbrukning jämfört en passivt dämpad skotare. En analys visar också att en ökad hastighet från 0,84 m/s till 1 m/s ökar energiförbrukningen med cirka 10% .

Reinforcement learning is conceptually based on an agent learning through interaction with its environment. This trial-and-error learning method makes the process prone to situations in which the agent is stuck in a dead-end, from which it cannot keep learning. This thesis studies a method to diminish the risk that a wheeled inverted pendulum, or WIP, falls over during training by having a Qlearning based agent estimate a PID controller before training it on the balance problem. We show that our approach is equally stable compared to a Q-learning agent without estimation training, while having the WIP falling over less than half the number of times during training. Both agents succeeds in balancing the WIP for a full hour in repeated tests. / Reinforcement learning baseras på en agent som lär sig genom att interagera med sin omgivning. Denna inlärningsmetod kan göra att agenten hamnar i situationer där den fastnar och inte kan fortsätta träningen. I denna examensuppsats utforskas en metod för att minska risken att en självkörande robot faller under inlärning. Detta görs genom att en Q-learning agent tränas till att estimera en PID kontroller innan den tränar på balanseringsproblemet. Vi visar att vår metod är likvärdigt stabil jämfört med en Q-learning agent utan estimeringsträning. Under träning faller roboten färre än hälften så många gånger när den kontrolleras av vår metod. Båda agenterna lyckas balansera roboten under en hel timme.

The topic of this thesis is base isolation. The purpose of this thesis is to offer a relative understanding of the seismic performance enhancements that a typical 12-story steel office building can achieve through the implementation of base isolation technology. To reach this understanding, the structures of a fixed-base office building and a base-isolated office building of similar size and layout are designed, their seismic performance is compared, and a cost-benefit analysis is completed. The base isolation system that is utilized is composed of Triple Friction Pendulum (TFP) bearings. The work of this thesis is divided into four phases. First, in the building selection phase, the structural systems (SMF and SCBF), layout, location (San Diego, CA), and design parameters of the buildings are selected. Then, in the design phase, each structure is designed using modal response spectrum analysis in ETABS. In the analysis phase, nonlinear time history analyses at DBE and MCE levels are conducted in PERFORM-3D to obtain the related floor accelerations and interstory drifts. Finally, in the performance assessment phase, probable damage costs are computed using fragility curves and FEMA P-58 methodology in PACT. Damage costs are computed for each building and seismic demand level and the results are compared.

Friction is one of the essential aspects of pavement performance and safety. Unfortunately, the rate at which the friction data are being collected exceeds the rate at which the data can be proficiently analyzed. Furthermore, the Utah Department of Transportation (UDOT) lacks long-term trend analysis for the many years of locked-wheel skid trailer (LWST) data collected in Utah. In addition, UDOT is missing a statistically adequate correlation equation between friction-testing devices. Likewise, only one method is used in Utah to prequalify aggregates for use in pavements. Finally, there has not been an investigation of the potential use of lithium silicate solution in Utah as a hardening agent to decrease the rate of friction loss. This research consists of five objectives. The first objective was to investigate pavement friction factors that influence skid resistance; methods of measuring skid resistance in the laboratory and the field, including correlations between test results; methods of evaluating aggregate sources; and methods of enhancing skid resistance of asphalt pavements through a comprehensive literature review on these subjects. The second objective was to investigate temporal trends in skid numbers measured using the LWST on Utah highways with different surface treatment types. The third objective was to develop a three-way correlation between the skid number measured with the LWST in the field, the British pendulum number measured with the British pendulum tester (BPT) in the field, and the polish value measured with the BPT in the laboratory. The fourth objective was to investigate selected performance-related properties of aggregates used to produce surface treatments at several field sites representing Utah conditions. The fifth objective was to examine the potential benefits of lithium silicate treatment for improving the resistance of aggregates to polishing. The scope of the research for the five objectives included statistical analysis, field testing, and laboratory experimentation. The findings include, first, a literature review that identified four critical deficiencies in Utah’s friction-related literature, which formed the basis of the remaining four objectives. Second, a statistical analysis of 9 years of LWST data indicated above-average skid values across Utah’s pavement network. Third, correlations were evaluated for multiple friction-testing devices. Fourth, X-ray diffraction testing methods were found to compare favorably to the accelerated polish test. Fifth and finally, the effects of lithium silicate solution on polish-susceptible aggregates were documented. This research has substantially advanced the body of knowledge on pavement friction testing and improving the resistance of aggregates to polishing in Utah through laboratory and field experimentation

Deep Reinforcement Learning (RL) has received much attention in recent years. This thesis investigates how reward functions, environment termination conditions, Neural Network (NN) architectures, and the type of the deep RL algorithm aect the performance for continuous control tasks. To this end, the Furuta pendulum swing-up task is adopted as the primary benchmark, since it oers low input- and state-dimensionality without being trivial. Focusing on model-free algorithms, the results indicate that DDPG, an actorcritic algorithm, performs significantly better than other algorithms. They also suggest that larger NN architectures may benefit performance in some instances. Comparing reward functions, Potential Based Reward Shaping (PBRS) applied to a sparse reward signal shows promising results compared to a reward function of previous work, and combining PBRS with large negative rewards for terminations due to unwanted behavior seems to improve performance for some algorithms. However, although designs such as PBRS can improve performance they are shown to not be necessary to achieve adequate performance, and the same applies to environment terminations upon unwanted behavior. Attempting to apply a DDPG agent trained in a simulator to a physical Furuta pendulum results in performance that closely resembles what is observed in the simulator for certain training seeds. The results and test suite of this thesis are available on GitHub and should hopefully help inspire future research in environment design and NN architectures for deep RL. Specifically, future work may investigate whether extensive parametertuning alters the results. / Djup förstärkningsinlärning har fått mycket uppmärksamhet de senaste åren. Detta arbete undersöker hur belöningsfunktioner, miljöers termineringsvillkor, neurala nätverksarkitekturer, och typen av djup förstärkningsinlärningsalgoritm påverkar prestandan för kontroll av kontinuerliga system. För att uppnå detta används uppsvängning av Furuta-pendeln som primärt referensproblem, ty det har få indata- och tillståndsdimensioner utan att vara trivialt. Fokus riktas mot modellfria algoritmer, där resultaten indikerar att DDPG, en aktörkritisk algoritm, presterar signifikant bättre än andra algoritmer. Resultaten indikerar också att större nätverksarkitekturer kan ge bättre prestanda i vissa fall. Vid jämförelse av belöningsfunktioner visar potentialbaseradbelöningsutformning (PBRS) applicerat på en gles belöningsfunktion lovande resultat jämfört med en belöningsfunktion från tidigare forskning, och kombinationen av PBRS med stora negativa belöningar för termineringar på grund av oönskat beteende verkar förbättra prestandan för vissa algoritmer. Dock, även om designer så som PBRS kan förbättra prestandan påvisas det att de inte är nödvändiga för att uppnå adekvat prestanda, och detsamma gäller miljötermineringar vid oönskat beteende. Försöket med applicering av en DDPG-agent tränad i en simulator på en fysisk Furuta-pendel resulterar i prestanda som nära efterliknar vad som uppnås i simulatorn för särskilda träningsfrön. Resultaten och testsviten för detta projekt finns tillgängliga på GitHub och kommer förhoppningsvis inspirera framtida forskning inom miljödesign och neurala nätverksarkitekturer för djup förstärkningsinlärning. Specifikt så kan framtida arbeten utreda huruvida utförlig parameterjustering påverkar resultaten.