The problem of synthesizing controllers for motion planning of multi-agent systems under Linear Temporal Logic (LTL) high-level specifications has been of great interest and has been widely studied over the last years. However, LTL cannot handle time constraints as specifications. The time aspect would allow more complicated and specific tasks and it is therefore desirable to incorporate. This work aims to determine how control synthesis for a continuous linear system can be performed based on Metric Interval Temporal Logic (MITL), which is able to handle desired time constraints to high-level specifications. Firstly, a control design synthesis method for a single-agent, based on previous work within both the field of LTL and MITL is presented. Secondly, a control design synthesis method for multi-agent systems considering both local an global MITL specifications is presented. Extended simulations has been performed in MATLAB environment demonstrating the two proposed methodologies. The result shows that the methods guarantee that the MITL specifications are satisfied, for all cases for which a solution is found. / Problemet gällande regulator syntetisering for rörelse planering av fler-agents system under Line-ar Temporal Logic (Linjär Temporal Logik=LTL) hög-nivå specifikationer har varit av stort intresse och har studerats brett under de senaste åren. LTL kan emellertid inte hantera tidsbegränsningar som specifikationer. Tidsaspekten skulle tillåta mer komplicerade och specifika uppgifter. Det är därför önskvärt att inkorporera. Målet med det här arbetet är att fastställa hur regulator syntetisering för ett kontinuerligt, linjärt system kan utföras utgående från Metric Interval Temporal Logic (Metrisk Intervall Temporal Logic =MITL), en gren av Temporal Logik som kan hantera de önskvärda tidsbegränsningarna för högnivå specifikationer. Först presenteras en metod för att syntetisera regulatorer för en-agents system. Metoden är baserad på tidigare arbeten inom fälten LTL och MITL. Sedan presenteras en metod för att syntetisera regulatorer för fler-agents system som ¨önskas uppfylla såväl lokala som globala MITL specifikationer. Utbredda simulationer har genomförts i MATLAB miljö för att demonstrera de två˚ föreslagna metoderna. Resultatet visar att metoderna garanterar att MITL specifikationerna är uppfyllda för alla fall för vilka en lösning hittas.

The trend toward the electrification of functionalities magnifies the part played by power electronic converters, which are continuously expected to deliver more power, at a lower cost, in a smaller volume and always with a high reliability. To this end, high switching frequency integrated converters, and especially Printed-Circuit-Board (PCB) –integrated devices, could embody an interesting paradigm. This however raises new challenges, in particular regarding the need for low-profile, planar, components. This study focuses on a structure of magnetic component made of a PCB placed between two plates of ferro- or ferri-magnetic material. Concentric spirals are printed on the PCB to form the winding(s) of the device. The plates of magnetic material decrease the reluctance of the magnetic path, whilst improving the Electro-Magnetic Compatibility (EMC) of the device by acting as a magnetic shield. A multi-physical modelling of this structure is described in this report. Both the magneto-static and magneto-dynamic are characterised, the electrostatics of the device is estimated, expressions of the core and copper losses are derived, thermal is assessed and some considerations on EMC are drawn.  Finally, a prototype is manufactured and characterised. The measured results are confronted with those obtained using Finite Elements Analysis (FEA) and the derived modelling. An ideality factor is introduced as an estimator of the correlation between the results simulated using FEA and those of the modelling. It appears that, in the somewhat restricted domain of interest, the correlation is satisfying. As for the physical measurements, they differ from the calculated and simulated results mostly due to the limited accuracy of the modelling and to the chosen dimensioning not being ideal enough for the derived modelling to give proper results. / Trenden att elektrifiera diverse funktionaliteter förstärker betydelsen av effektelektroniken, vilken förväntas att kunna hantera högre effekter till lägre kostnad och mindre volym, och med en hög tillförlitlighet. Mot bakgrund av detta är integrerade effektomvandlare med hög switchfrekvens, och speciellt komponenter som integreras i kretskortet, ett mycket intressant utvecklingsområde. Detta innebär emellertid nya utmaningar, speciellt beträffande behovet av nya planara komponenter med låg bygghöjd. Detta arbete fokuserar på en struktur för magnetiska komponenter som bygger på ett kretskort som placeras mellan två skikt av ferro- eller ferrimagnetiska material. Koncentriska spiraler realiseras som lindningar på kretskortet. Skikten med magnetiskt material minskar reluktansen i den magnetiska kretsen, samtidigt som den förbättrar den elektromagnetiska kompatibiliteten (EMC) genom att verka som skärmar. En multifysik-modell av denna struktur beskrivs i denna rapport. Med denna undersöks magnetostatiska, magnetodynamiska och elektrostatiska egenskaper. Dessutom härleds uttryck för kärn- och lindningsförluster samt termiska storheter. Även vissa slutsatser meträffande EMC dras. Slutligen tillverkas och karakteriseras en prototyp. De uppmätta värdena jämförs med värden beräknade med finita-elementanalys och värden erhållna med hjälp av den utvecklade modellen. En idealitetsfaktor införs för kvantifiering av korrelationen mellan värden från modellen och finita-elementanalysen. Det förefaller som om correlationen är tillfredställande inom det något restriktiva arbetsområdet. När det gäller de experimentella resultaten avviker de från de beräknade och simulerade värdena i första hand pga den begränsade noggrannheten i modelleringen och den valda dimensioneringen, som inte var tillräckligt ideal för att modelleringen skulle kunna ge tillräckligt goda resultat.

We consider the modelling and optimal control of energy storage systems, in this study a Vanadium Redox Flow Battery. Such a battery can be introduced in the electrical grid to be charged when demand is low and discharged when demand is high, increasing the overall efficiency of the network while reducing costs and emission of greenhouse gases. The model of the battery proposed in this study is less complex than the majority of models on batteries and energy storage systems found in literature, but still accurate enough to capture the core aspects of the behaviour of the battery. Estimation methods are discussed for determining unknown parameters in the model using experimental data and the model is evaluated against additional measurements of the physical system showing good fit. The purpose of having a simple model is that the resulting optimal control problem is easier to solve. Operation of the battery with respect to several different objectives are discussed and that of minimizing the vertical net load and maximizing profit are solved for open loop control and receding horizon control over the time span of one day. Two performance metrics are proposed for determining the gain of operating an electrical network with or without the battery. The results show that the introduction of storage systems indeed helps in increasing profit and reducing emissions.

The detection of anomalies occurring in a network is of great importance. Networks need to guarantee performance to their users as well as their security. The detection of anomalies needs to be done as quickly as possible to provide an appropriate response to the threat (block the traffic of an attack, bring additional servers to answer a high demand). The thesis answers the question: can the detection of anomalies be done by using active monitoring? Active monitoring is done by sending probe packets on a network to evaluate the state of the traffic. Active monitoring generates an additional traffic on the network. Several metrics can be measured but not all are interesting for the detection of anomalies. Several detection methods have been developed over the years, and several categories exist. The supervised and semi-supervised algorithms need labeled data while unsupervised algorithms do not. The thesis develops a solution using an active tool measuring the available bandwidth of a network and a statistical detection algorithm based on change point detection. The solution has been tested on a controlled testbed against Denial of Service attacks (DoS) and shows promising results against them, but hasn’t been able to detect network scanning.

No description available.

Calibration of particle detectors for space plasma sensors is an essential part of instrument development process. The Jovian plasma Dynamics and Composition analyzer (JDC) is one of the two instruments built by the Swedish Institute of Space Physics, for the Europen Space Agency’s JUICE (JUpiter ICy moon Explorer) mission. RATEX-J (RAdiation Test EXperiment for JUICE) is a project aiming to perform radiation tests to characterize the sensors and test the anti-coincidence system of JDC. Particle detectors operate in high quality vacuum only, hence a vacuum system is imperative for these tests.This report focuses on the vacuum system designed for the RATEX-J experiment. After the requirements were established, a DN 160 ISO-K four way cross was selected as a vacuum tank, then each wall was designed to accommodate the needed appliances and connectors. Finally a support structure was built to meet mobility and stability requirements.The system performance during the first radiation campaign was satisfactory. After concluding the test results, a few alterations were requested to adapt the system for the next test session.Finally a larger vacuum system and a motorized manipulator design is discussed, which are planned for radiation test of the complete JDC instrument.

In this work, a trajectory tracking controller structure for a systemcomposed of a load and an aerial vehicle, which are connected by a rope, is presented. The main goal is to design a controller that guarantees that the load tracks a desired trajectory asymptotically. The load is modeled as a point mass while the aerial vehicle is assumed to be fully actuated. The load and the aerial vehicle are connected via a ropelike structure of fixed length that can only withstand tensile forces along itsorientation. This system can be transformed into a form that resembles the form of systems used to describe underactuated aerial vehicles. A backstepping procedure is used to design a nonlinear control law whichis based on a controller for a double integrator. A bounded controllaw for double integrators is developed such that the position of the load converges to the desired trajectory for initial positions that are arbitrarily far away from the desired trajectory. The proposed controller structure has been verified both in simulations and with experiments using quadcopters at the Smart Mobility Lab at KTH.

Thales Alenia Space is currently developing an innovative satellite platform with a two-phase fluid loop on board. The company uses two softwares in order to create models that are meaningful regarding physics: Amesim and e-Therm. The first one is used to describe fluid behavior in a circuit, while the second one is used to simulate conductive and radiative heat transfers in a satellite model but cannot be used to model fluid components. The interconnection of both software through a cosimulation module would allow to create dynamic models that would take into account both physical aspects. My internship was initially supposed to deal with the debugging of that cosimulation module, however a delay for its delivery forced me to reorient myself. First, I used e-Therm to model 3D heat pipes profiles, before working on the thermal model of a Spacebus Neo satellite in order to check the validity of the design choices that were kept for that satellite. / Thales Alenia Space développe actuellement une plateforme de satellite innovante embarquant une boucle fluide diphasique. Dans le but de créer des modèles de satellites représentatifs de la réalité physique, l’entreprise dispose de deux logiciels : Amesim et e-Therm. Le premier décrit très bien les circuits fluidiques, tandis que le second modélise les échanges conductifs et radiatifs dans un modèle de satellite mais ne peut pas modéliser d’élément fluidique. La mise en relation de ces deux logiciels par un module de cosimulation permettrait de créer des modèles dynamiques prenant en compte les deux aspects physiques. Mon stage s’inscrivait dans le développement de ce module de cosimulation, mais un retard sur sa livraison m’a forcé à me réorienter. Mon travail a donc consisté à utiliser e-Therm pour modéliser des profilés de caloducs dans un premier temps, avant de travailler sur un modèle de satellite Spacebus Neo dans le but de vérifier si les choix de design qui avaient été retenus étaient valides.

Autonomous driving vehicles introduce challenging research areas combining differ-ent disciplines. One challenge is the detection of obstacles with different sensors and the combination of information to generate a comprehensive representation of the environment, which can be used for path planning and decision making.The sensor fusion is demonstrated using two Velodyne multi beam laser scanners, but it is possible to extend the proposed sensor fusion framework for different sensor types. Sensor fusion methods are highly dependent on an accurate pose estimate, which can not be guaranteed in any case. A fault tolerant sensor fusion based on the Dempster Shafer theory to take the uncertainty of the pose estimate into account is discussed and compared using an example, although not implemented on the test vehicle.Based on the fused occupancy grid map, dynamic obstacles are tracked to give a velocity estimate without the need of any object or track association methods. Ex-periments are carried out on real world data as well as on simulated measurements, for which a ground truth reference is provided.The occupancy grid mapping algorithm runs on central- and graphical-processing units, which allows to give a comparison between the two approaches and to stress out which approach is preferably used, depending on the application. / Självkörande bilar har lett till flera intressanta forskningsområden som kombinerar manga olika discipliner. En utmaning är att ge fordonet en sorts ¨ögon. Genom att använda ytterligare sensorer och kombinera data frän samtliga så kan man detektera hinder i fordonets väg. Detta kan naturligtvis användas för att förbättra fordonets planerade rutt och därmed också minska klimatpåverkan. Här används två sammankopplade Velodyne laserstrålsensorer för att undersöka detta Närmare, men det går också att utöka antalet sensorer ytterligare. Sammanlänkningen av sensorer är mycket känslig och kräver därför exakta koordinater, vilket inte alltid kan garanteras. Därför utreds istället om en sensor baserad på Dempster Shaferteorin kan användas för att hantera fel och osäkerheter. Denna används dock inte i testfordonet. Baserat på en sammanvägd kartbild över upptagna och fria områden (occupancy grid mapping) kan objekt och hinder i rörelse följas för att uppskatta fordonets hastighet utan att metoder för objekt- eller banidentifiering behöver användas. Experiment har utförts på verklig data. Dessutom används simulerade mätningar där en sann grundreferens används. Algoritmen som används för occupancy-kartan använder sig av central- och grafik-processorenheter, vilket ger oss möjlighet att jämföra två metoder och finna den bäst fungerande metoden för olika applikationer.

With the emergence of smartphones, data traÿc became the dominant application type in current mobile networks. This thesis deals with the comparison of the per-formance of di˙erent TCP congestion control algorithms when running in an LTE network. We developed an environment allowing for the simulation of the Linux im-plementation of di˙erent TCP variants and compared their performance in di˙erent scenarios. The results show that the loss-based variants manage to reach full link util-isation but create huge amount of delay whereas the delay-based mechanisms keep the delay under control but are not always able to fill the link. / På grund av utvecklingen av smartphones blev data den största källan av traffik i aktuella mobilnätverk. Den här uppsatsen handlar om jämförelsen av utförandet av olika TCP Congestion Control Algorithms när de används i LTE nätverk. Vi utveklade ett system som kan användas för att simulera Linux implementeringar av olika TCP versioner och jämföra deras utförande i olika situationer.