There is currently a high interest of replacing traditional wired networks with wireless technology. Wireless communications can provide several advantages for process industries with aspect to exibility, maintenance and installation. The WirelessHART protocol provides a standardized wireless technology for large automation networks that explore wireless communication. However, wireless networks introduce time delays and losses in the communication system, which denes requirements for designing estimators and controllers that can tolerate and compensate for the losses and delays. This thesis consists of several contributions. First, we develop tools for analyzing the delay and loss probabilities in WirelessHART networks with unreliable transmission links. For given network topology, routing and transmission schedule the developed tools can be used to determine the latency distributions of individual packets and quantify that a packet will arrive within a prescribed deadline. Secondly, we consider estimation and control when sensor and control messages are sent over WirelessHART networks. The network losses and latencies are modelled and compensated for by timevarying Kalman lters and LQG controllers. Both optimal controllers, of high implementation complexity, and simple suboptimal schemes are considered. The control strategies are evaluated on a simulation model of a flotation process in a Boliden mine where the wired sensors of the existent solution are replaced by a WirelessHART network scheduled for time-optimal data collection. Finally, we implement a WirelessHART-compliant sensor on a Tmote sky device and perform real experiments of wireless control on a water tank process.

Une étude d'un assemblage symétrique de trois 4-simplex en géométrie classique, de Regge et quantique. Nous étudions les propriétés géométriques et surtout la présence de courbure. Nous montrons que les géométries classique et de Regge de l'assemblage ont une courbure qui évolue en fonction de ses paramètres de bordure. Pour la géométrie quantique, une version euclidienne du modèle EPRL est utilisé avec une valeur pratique du paramètre Barbero-Immirzi pour définir l'amplitude de transition de l'ensemble et de ses composants. Un code C ++ est conçu pour calculer les amplitudes et étudier numériquement la géométrie quantique. Nous montrons qu'une géométrie classique, avec une courbure, émerge déjà à bas spin. Nous reconnaissons également l'apparition de configurations dégénérées et de leurs effets sur la géométrie attendue. / A study of symmetrical assembly of three euclidean 4-simplices in classical, Regge and quantum geometry. We study the geometric properties and especially the presence of curvature. We show that classical and Regge geometry of the assembly have curvature which evolves in function of its boundary parameters. For the quantum geometry, a euclidean version of EPRL model is used with a convenient value of the Barbero-Immirzi parameter to define the transition amplitude of the assembly and its components. A C++ code is design for compute the amplitudes and study numerically the quantum geometry. We show that a classical geometry, with curvature, emerges already at low spin. We also recognize the appearance of the degenerate configurations and their effects on the expected geometry.

Neste trabalho, duas técnicas de controle robusto (LQG/LTR e DML), aplicadas a um sistema elétrico de potência, são avaliadas através dos raios de estabilidade do sistema. As incertezas estruturadas do modelo nominal são consideradas nos dois controladores. Um conjunto de modelos é gerado considerando as combinações das incertezas paramétricas. Os valores singulares estruturados dos dois sistemas de controle são analisados. / In this work, two techniques of robust control (LQG/LTR) and LMI), applied to a power electric system, are available via stability radii of the system. The structured uncertainties of the nominal model are considered in both designs. A set of models is generated considering the combinations of the parametric uncertainties. The structured singular values of both systems are analysed.

Neste trabalho, duas técnicas de controle robusto (LQG/LTR e DML), aplicadas a um sistema elétrico de potência, são avaliadas através dos raios de estabilidade do sistema. As incertezas estruturadas do modelo nominal são consideradas nos dois controladores. Um conjunto de modelos é gerado considerando as combinações das incertezas paramétricas. Os valores singulares estruturados dos dois sistemas de controle são analisados. / In this work, two techniques of robust control (LQG/LTR) and LMI), applied to a power electric system, are available via stability radii of the system. The structured uncertainties of the nominal model are considered in both designs. A set of models is generated considering the combinations of the parametric uncertainties. The structured singular values of both systems are analysed.

High Redundancy Actuation (HRA) is a novel type of fault tolerant actuator. By comprising a relatively large number of actuation elements, faults in the elements can be inherently accommodated without resulting in a failure of the complete actuation system. By removing the possibility of faults detection and reconfiguration, HRA can provide high reliability and availability. The idea is motivated by the composition of human musculature. Our musculature can sustain damage and still function, sometimes with reduced performance, and even complete loss of a muscle group can be accommodated through kinematics redundancy, e.g. the use of just one leg. Electro-mechanical actuation is used as single element inside HRA. This thesis is started with modelling and simulation of individual actuation element and two basic structures to connect elements, in series and in parallel. A relatively simple HRA is then modelled which engages a two-by-two series-in-parallel configuration. Based on this HRA, position feedback controllers are designed using both classical and optimal algorithms under two control structures. All controllers are tested under both healthy and faults injected situations. Finally, a hardware demonstrator is set up based simulation studies. The demonstrator is controlled in real time using an xPC Target system. Experimental results show that the HRA can continuously work when one element fails, although performance degradation can be expected.

<p>The goal of this Master thesis has on the one hand been to create a simulation environment for robots, where not only the simulations in this work can be performed, but also similar robot simulations, and on the other hand to study regulators that can increase the possibilities of rejecting disturbances on an industrial robot. The robot model used has three degrees of freedom, corresponding to the three main links on ABB:s IRB7600. </p><p>The simulation environment is created using Matlab, Simulink, and Robotics Toolbox. It is constructed in a general way and can be used for simulations of other robot objects created in Robotics Toolbox. </p><p>A gain scheduling is added to the LQ-regulator, where the feedback matrix is varied depending on the arm position in the workspace. The gain scheduling is implemented along a reference path in the workspace, around which the simulations and frequency analyses are performed. </p><p>The regulators that have been tested are a PID-regulator, a gain scheduled LQ-regulator, a gain scheduled LQ-regulator with reconstructed states, and finally a fix LQ-regulator with reconstructed states. Results from the tests show that the gain scheduling LQ-regulator is the one that yields the best rejection of disturbances, but the corresponding regulator with reconstructed states, which is more realistic, performs nearly as well. The model based regulators are, however, significantly better when it comes to rejecting disturbances than the PID-regulator.</p>

Many traffic incidents are due to the driver’s lack of attention, resulting in dangerous lane departures, either sliding off theroad or into the oppose lane. These kinds of incidents often have serious outcomes, which has led to much effort being concentrated on preventing or lessening the damages when the incident is already a fact, for example by installing safety belts and air bags. These measures may be considered to be acts of so-called passive safety. Active safety on the other hand, means that the safety systems intervene before the incidents have occurred. Lane Keeping Aid (LKA), which has been developed and implemented in this master thesis project, is a system designed to support the driver in the lateral axis in situations when unwanted lane departure is an evident risk. To be able to determine when the system should intervene and support the driver, information regarding how the driver handles the vehicle, along with the vehicle’s position and direction in the lane, is essential. The car’s position may be obtained by installing a camera in the vehicle. The information needed regarding other things, e.g. the car’s position in relation to the lane, is obtained by using a Kalmanfilter, which is based on a physically developed model, and which estimates the mentioned distances. Based on measurements and estimated values, it is possible for the LKA system to calculate an assistance torque, aimed at decreasing the lateral deviation from the centre of the lane. An electric power steering, instead of a conventional hydraulic steering servo is then used to produce the torque. The LKA system has been developed in a simulation environment using Simulink before being implemented, in order to monitor the function of the system before beginning actual testdrives. Furthermore, real measurement data given at driving with the test vehicle has been used to adjust and test the function. The results from the project’s first phase, in the simulation environment, show that the estimated values from the Kalmanfilter correlates well with real test data. Simulations with real measurement data show that the system functions as intended. Finally, it may also be mentioned, that the system has yet not been fully tested in a vehicle equipped with an electric power steering, which ought to be included in future development of the system.

The goal of this Master thesis has on the one hand been to create a simulation environment for robots, where not only the simulations in this work can be performed, but also similar robot simulations, and on the other hand to study regulators that can increase the possibilities of rejecting disturbances on an industrial robot. The robot model used has three degrees of freedom, corresponding to the three main links on ABB:s IRB7600. The simulation environment is created using Matlab, Simulink, and Robotics Toolbox. It is constructed in a general way and can be used for simulations of other robot objects created in Robotics Toolbox. A gain scheduling is added to the LQ-regulator, where the feedback matrix is varied depending on the arm position in the workspace. The gain scheduling is implemented along a reference path in the workspace, around which the simulations and frequency analyses are performed. The regulators that have been tested are a PID-regulator, a gain scheduled LQ-regulator, a gain scheduled LQ-regulator with reconstructed states, and finally a fix LQ-regulator with reconstructed states. Results from the tests show that the gain scheduling LQ-regulator is the one that yields the best rejection of disturbances, but the corresponding regulator with reconstructed states, which is more realistic, performs nearly as well. The model based regulators are, however, significantly better when it comes to rejecting disturbances than the PID-regulator.

Linearized stability analysis methodologies, system identification algorithms and optimal control approaches that are applicable to large scale, flexible multibody dynamic systems are presented in this thesis. For stability analysis, two classes of closely related algorithms based on a partial Floquet approach and on an autoregressive approach, respectively, are presented in a common framework that underlines their similarity and their relationship to other methods. The robustness of the proposed approach is improved by using optimized signals that are derived from the proper orthogonal modes of the system. Finally, a signal synthesis procedure based on the identified frequencies and damping rates is shown to be an important tool for assessing the accuracy of the identified parameters; furthermore, it provides a means of resolving the frequency indeterminacy associated with the eigenvalues of the transition matrix for periodic systems. For system identification, a robust algorithm is developed to construct subspace plant models. This algorithm uniquely combines the methods of minimum realization and subspace identification. It bypasses the computation of Markov parameters because the free impulse response of the system can be directly computed in the present computational environment. Minimum realization concepts were applied to identify the stability and output matrices. On the other hand, subspace identification algorithms construct a state space plant model of linear system by using computationally expensive oblique matrix projection operations. The proposed algorithm avoids this burden by computing the Kalman filter gain matrix and model dependency on external inputs in a small sized subspace. Balanced model truncation and similarity transformation form the theoretical foundation of proposed algorithm. Finally, a forward innovation model is constructed and estimates the input-output behavior of the system within a specified level of accuracy. The proposed system identification algorithms are computationally inexpensive and consist of purely post processing steps that can be used with any multi-physics computational tool or with experimental data. Optimal control methodologies that are applicable to comprehensive large-scale flexible multibody systems are presented. A classical linear quadratic Gaussian controller is designed, including subspace plant identification, the evaluation of linear quadratic regulator feedback gain and Kalman filter gain matrices and online control implementation.

The architectures of the control devices in active control algorithm are an important fact in civil structural buildings. Traditional research has limitations in finding the optimal architecture of control devices such as using predefined numbers or locations of sensors and dampers within the 2-and 3-dimensional (3-D) model of the structure. Previous research using single-objective optimization only provides limited data for defining the architecture of sensors and control devices. The Linear Quadratic Gaussian (LQG) control algorithm is used as the active control strategy. The American Society of Civil Engineers (ASCE) control benchmark building definition is used to develop the building system model. The proposed gene manipulation genetic algorithm (GMGA) determines the near-optimal Pareto fronts which consist of varying numbers and locations of sensors and control devices for controlling the ASCE benchmark building by considering multi-objectives such as interstory drift and minimizing the number of the control devices. The proposed GMGA reduced the central processing unit (CPU) run time and produced more optimal Pareto fronts for the 2-D and 3-D 20-story building models. Using the GMGA provided several benefits: (1) the possibility to apply any presuggested multi-objective optimization mechanism; (2) the availability to perform a objective optimization problem; (3) the adoptability of the diverse encoding provided by the GA; (4) the possibility of including the engineering judgment in generating the next generation population by using a gene creation mechanisms; and (5) the flexibility of the gene creation mechanism in applying and changing the mechanism dependent on optimization problem. The near-optimal Pareto fronts obtained offer the structural engineer a diverse choice in designing control system and installing the control devices. The locations and numbers of the dampers and sensors in each story are highly dependent on the sensor locations. By providing near-Pareto fronts of possible solutions to the engineer that also consider diverse earthquakes, the engineer can get normalized patterns of architectures of control devices and sensors about random earthquakes.