Dans ce mémoire de thèse nous présentons une méthodologie pour le diagnostic et la surveillance en ligne des éoliennes de l'état de l'ensemble machine électrique onduleur à partir de la mesure du champ magnétique continu. Ce travail est scindé en deux parties complémentaires. Dans la première partie, nous donnons quelques contributions concernant l'estimation de l'état et des paramètres pour certaines classes de systèmes non linéaires. Ces mêmes estimateurs ont été validés en simulation et sur un banc d'essai. La deuxième partie porte sur la mise en œuvre et la conception de deux bancs d'essai utilisés pour l'étude des défauts dans un système double alimentation MADA et un ensemble moteur électrique onduleur classique pour l'analyse des défauts du convertisseur. L'identification des paramètres et la technique de contrôle du système en boucle fermé est introduite pour l'application MADA. De plus, les capteurs virtuels sont conçus pour estimer la vitesse mécanique qui est utilisée pour calculer les fréquences caractéristiques défectueuses, et le signal de couple mécanique qui exerce une influence sur l'amplitude de certaines des signatures défectueux typiquement. Finalement, sur la base de la mise en équation théorique des signaux électriques avec défauts puis des simulations correspondantes, une comparaison avec l'expérimentation est réalisée pour valider la technique proposée. Ces travaux montrent l'avantage d'une telle technique dans la simplicité de l'instrumentation mis en œuvre / This doctoral thesis presents a methodology for the online condition monitoring of the electrical power drive in wind energy systems based on the local measurement of the DC-bus magnetic field. The work is divided into two complementary parts. In the first part, some contributions related to the estimation of the state and parameters for certain classes of nonlinear systems are provided. The estimators have been validated in simulation and on test benches. The second part focuses on the implementation and control design of two benchmarks used to study defects in a doubly-fed induction generator (DFIG) system and an induction motor power drive. In the former benchmark, the parameters identification of the induction machine and the controller design of the DFIG system are carried out. For the latter test bench, the notable work is to build an induction machine drive for the purpose of fault investigation, in which a PWM generator is developed to control and create the fault of the converter. Furthermore, virtual sensors are designed to estimate mechanical speed that is used to calculate the characteristic frequencies, and mechanical torque signal that has influence on the amplitude of some typical fault signatures. Finally, based on the theoretical aspect of the selected faults, the relevant simulations are developed and experiments are implemented on the benchmarks in order to validate the proposed technique. It has been shown that the diagnostic relying on the magnetic field measurement is feasible and offers various advantages such as simplicity and cost-effectiveness

With the increased performance and reduced cost of cameras, the robotics community has taken great interest in estimation and control algorithms that fuse camera data with other sensor data.In response to this interest, this dissertation investigates the algorithms needed for robust guidance, navigation, and control of fixed-wing and multirotor aircraft applied to target estimation and circumnavigation.This work begins with the development of a method to estimate target position relative to static landmarks, deriving and using a state-of-the-art EKF that estimates static landmarks in its state.Following this estimator, improvements are made to a nonlinear observer solving part of the SLAM problem.These improvements include a moving origin process to keep the coordinate origin within the camera field of view and a sliding window iteration algorithm to drastically improve convergence speed of the observer.Next, observers to directly estimate relative target position are created with a circumnavigation guidance law for a multirotor aircraft.Taking a look at fixed-wing aircraft, a state-dependent LQR controller with inputs based on vector fields is developed, in addition to an EKF derived from error state and Lie group theory to estimate aircraft state and inertial wind velocity.The robustness of this controller/estimator combination is demonstrated through Monte Carlo simulations.Next, the accuracy, robustness, and consistency of a state-of-the-art EKF are improved for multirotors by augmenting the filter with a drag coefficient, partial updates, and keyframe resets.Monte Carlo simulations demonstrate the improved accuracy and consistency of the augmented filter.Lastly, a visual-inertial EKF using image coordinates is derived, as well as an offline calibration tool to estimate the transforms needed for accurate, visual-inertial estimation algorithms.The imaged-based EKF and calibrator are also shown to be robust under various conditions through numerical simulation.

A neutralizacao de pH e utilizada nas industrias, para garantir o descarte seguro de euentes. As plantas de neutralizacao de pH sao um problema complexo de controle, visto que a planta segue um modelo nao-linear e apresenta caractersticas variantes no tempo, o que demanda sua correta modelagem para o projeto de sistemas de controle ecientes. No entanto, a teoria referente a modelagem de pH nao e facilmente aplicada na pratica, resultando frequentemente em modelos que nao predizem corretamente o comportamento dinamico da planta. O primeiro objetivo deste trabalho foi modelar matematicamente a Planta Piloto de Neutralizacao de pH do Laboratorio de Controle de Processos Industriais (LCPI), utilizando uma metodologia que possa ser aplicada para obter o modelo matematico de outras plantas de neutralizacao de pH. Inicialmente a Planta Piloto de Neutralizacao de pH do LCPI foi modelada de acordo com a abordagem fenomenologica, utilizando-se os princpios de conservacao de massa, da eletroneutralidade e os conceitos de equilbrio qumico. Em seguida, o modelo foi ajustado aos dados experimentais do processo (abordagem emprica), utilizando-se curvas de titulacao dos inuentes e distribuicoes de tempos de residencia do reator. Atraves de experimentos, vericou-se que o modelo representou, de forma satisfatoria, a dinamica real da Planta Piloto de Neutralizacao de pH do LCPI. Ademais, este modelo foi utilizado para alcancar o segundo objetivo deste trabalho: projetar um sistema de controle de pH, o qual foi composto por um observador nao-linear e um controlador baseado em modelo. Esta estrutura de controle foi testada experimentalmente, onde certicou-se que os requisitos de controle foram satisfeitos. / The pH neutralization is used in industry to discard properly the wastewater, ensuring the environment preservation. The pH neutralization is a complex control problem, as the model of the plant presents a strong nonlinearity and time varying characteristics, which demands a proper modeling in order to design ecient control systems. However, the application of the theory related to pH modeling is not a trivial task and may result in models that can not predict the plant dynamics. The rst objective of this research was to model the pH Neutralization Pilot Plant, of the Laboratory of Industrial Processes Control (LCPI), using a methodology that could be replicated to model other pH neutralization plants. Initially, the pH Neutralization Pilot Plant was modeled with the phenomenological approach, utilizing rst principles, such as the mass conservation, electroneutrality and chemical equilibrium. Moreover, the model was adjusted to represent the process observed data (empirical approach), as its titration curves of the inuent streams and its reactor residence time distribution. Through experiments, it was veried that the model could represent adequately the real process dynamics. Furthermore, this model was used to achieve the second objective of this research: to design a pH control system, which was composed of a nonlinear observer and a modelbased control. This control structure was tested experimentally, ensuring that the control requirements were satised.

The PhD presents a combined approach to improving individual car efficiency. An optimal observer, the Extended Kalman Filter, is used to create an efficiency model for the car. Particular attention was paid to handling the asynchronous and redundant nature of the measurement data. A low-cost sensor suite developed to measure data is described. This sensor suite was installed on multiple vehicles to good success. It employsan accelerometer, gps, fuel injector timer, and Vss input to measure all the data necessary to reconstruct the car's state. This observer and sensor suite can be used as the base for any study which requires car efficiency maps, allowing research to proceed without manufacturer supplied data. Once the efficiency map is found, it is then curve-fitted in order to reduce model complexity. The simplified model is then used as a basis for optimal control through Pontryagin's Maximum Principle. Real-world test results are given, both for efficiency mapping, and for optimal control. Detailed discussion of the observer and controller is presented, in order to ease understanding and save implementation time

A neutralizacao de pH e utilizada nas industrias, para garantir o descarte seguro de euentes. As plantas de neutralizacao de pH sao um problema complexo de controle, visto que a planta segue um modelo nao-linear e apresenta caractersticas variantes no tempo, o que demanda sua correta modelagem para o projeto de sistemas de controle ecientes. No entanto, a teoria referente a modelagem de pH nao e facilmente aplicada na pratica, resultando frequentemente em modelos que nao predizem corretamente o comportamento dinamico da planta. O primeiro objetivo deste trabalho foi modelar matematicamente a Planta Piloto de Neutralizacao de pH do Laboratorio de Controle de Processos Industriais (LCPI), utilizando uma metodologia que possa ser aplicada para obter o modelo matematico de outras plantas de neutralizacao de pH. Inicialmente a Planta Piloto de Neutralizacao de pH do LCPI foi modelada de acordo com a abordagem fenomenologica, utilizando-se os princpios de conservacao de massa, da eletroneutralidade e os conceitos de equilbrio qumico. Em seguida, o modelo foi ajustado aos dados experimentais do processo (abordagem emprica), utilizando-se curvas de titulacao dos inuentes e distribuicoes de tempos de residencia do reator. Atraves de experimentos, vericou-se que o modelo representou, de forma satisfatoria, a dinamica real da Planta Piloto de Neutralizacao de pH do LCPI. Ademais, este modelo foi utilizado para alcancar o segundo objetivo deste trabalho: projetar um sistema de controle de pH, o qual foi composto por um observador nao-linear e um controlador baseado em modelo. Esta estrutura de controle foi testada experimentalmente, onde certicou-se que os requisitos de controle foram satisfeitos. / The pH neutralization is used in industry to discard properly the wastewater, ensuring the environment preservation. The pH neutralization is a complex control problem, as the model of the plant presents a strong nonlinearity and time varying characteristics, which demands a proper modeling in order to design ecient control systems. However, the application of the theory related to pH modeling is not a trivial task and may result in models that can not predict the plant dynamics. The rst objective of this research was to model the pH Neutralization Pilot Plant, of the Laboratory of Industrial Processes Control (LCPI), using a methodology that could be replicated to model other pH neutralization plants. Initially, the pH Neutralization Pilot Plant was modeled with the phenomenological approach, utilizing rst principles, such as the mass conservation, electroneutrality and chemical equilibrium. Moreover, the model was adjusted to represent the process observed data (empirical approach), as its titration curves of the inuent streams and its reactor residence time distribution. Through experiments, it was veried that the model could represent adequately the real process dynamics. Furthermore, this model was used to achieve the second objective of this research: to design a pH control system, which was composed of a nonlinear observer and a modelbased control. This control structure was tested experimentally, ensuring that the control requirements were satised.

Nous présentons un système de tracking optique-inertiel qui consiste en deux caméras stationnaires et une Sensor Unit avec des marqueurs optiques et une centrale inertielle. La Sensor Unit est fixée sur l'objet suivi et sa position et orientation sont déterminées par un algorithme de fusion de données. Le système de tracking est destiné à asservir un outil à main dans un système de chirurgie naviguée ou assistée par ordinateur. L'algorithme de fusion de données intègre les données des différents capteurs, c'est-à-dire les données optiques des caméras et les données inertielles des accéléromètres et gyroscopes. Nous présentons différents algorithmes qui rendent possible un tracking à grande bande passante avec au moins 200Hz avec des temps de latence bas grâce à une approche directe et des filtres dits invariants qui prennent en compte les symmétries du système. Grâce à ces propriétés, le système de tracking satisfait les conditions pour l'application désirée. Le système a été implementé et testé avec succès avec un dispositif expérimental. / We present an optical-inertial tracking system which consists of two stationary cameras and a Sensor Unit with optical markers and an inertial measurement unit (IMU). This Sensor Unit is attached to the object being tracked and its position and orientation are determined by a data fusion algorithm.The tracking system is to be used for servo-controlling a handheld tool in a navigated or computer-assisted surgery system.The data fusion algorithm integrates data from the different sensors, that is optical data from the cameras and inertial data from accelerometers and gyroscopes. We present different algorithms which ensure high-bandwidth tracking with at least 200Hz with low latencies by using a direct approach and so-called invariant filters which take into account system symmetries. Through these features, the tracking system meets the requirements for being used in the desired application.The system was successfully implemented and tested with an experimental setup.

Ce mémoire de thèse présente une méthode permettant d'améliorer laconduite automobile. Le filtre de Kalman étendu est utilisé pour identifierun modèle de la voiture. Ce filtre est particulièrement étudié afin de prendreen compte la redondance des informations et leur mesure asynchrone.Un ensemble cohérent et bon marché de capteurs - incluant accéléromètres,GPS, temps d'ouverture des injecteurs et vitesse - a été développé et installédans plusieurs véhicules. Ces mesures sont utilisées afin de reconstituer lafonction d'efficacité du moteur. Cette méthodologie peut-être utilisée pourtoute étude requérant la connaissance de cette fonction.La fonction d'efficacité est approchée par une fonction polynomiale etle modèle obtenu est la base d'une optimisation utilisant le principe dumaximum de Pontryagin.Les résultats des tests en condition réelle sont donnés et montrent l'efficicacité de l'observateur et du contrôleur / The PhD presents a combined approach to improving individual car efficiency. An optimal observer, the Extended Kalman Filter, is used to create an efficiency model for the car. Particular attention was paid to handling the asynchronous and redundant nature of the measurement data. A low-cost sensor suite developed to measure data is described. This sensor suite was installed on multiple vehicles to good success. It employsan accelerometer, gps, fuel injector timer, and Vss input to measure all the data necessary to reconstruct the car's state. This observer and sensor suite can be used as the base for any study which requires car efficiency maps, allowing research to proceed without manufacturer supplied data. Once the efficiency map is found, it is then curve-fitted in order to reduce model complexity. The simplified model is then used as a basis for optimal control through Pontryagin's Maximum Principle. Real-world test results are given, both for efficiency mapping, and for optimal control. Detailed discussion of the observer and controller is presented, in order to ease understanding and save implementation time

The main body of this thesis consists of six appended papers. In the  first two, different  cooperative surveillance problems are considered. The second two consider different aspects of the trajectory planning problem, while the last two deal with observer design for mobile robotic and Euler-Lagrange systems respectively.In Papers A and B,  a combinatorial optimization based framework to cooperative surveillance missions using multiple Unmanned Ground Vehicles (UGVs) is proposed. In particular, Paper A  considers the the Minimum Time UGV Surveillance Problem (MTUSP) while Paper B treats the Connectivity Constrained UGV Surveillance Problem (CUSP). The minimum time formulation is the following. Given a set of surveillance UGVs and a polyhedral area, find waypoint-paths for all UGVs such that every point of the area is visible from  a point on a waypoint-path and such that the time for executing the search in parallel is minimized.  The connectivity constrained formulation  extends the MTUSP by additionally requiring the induced information graph to be  kept recurrently connected  at the time instants when the UGVs  perform the surveillance mission.  In these two papers, the NP-hardness of  both these problems are shown and decomposition techniques are proposed that allow us to find an approximative solution efficiently in an algorithmic manner.Paper C addresses the problem of designing a real time, high performance trajectory planner for an aerial vehicle that uses information about terrain and enemy threats, to fly low and avoid radar exposure on the way to a given target. The high-level framework augments Receding Horizon Control (RHC) with a graph based terminal cost that captures the global characteristics of the environment.  An important issue with RHC is to make sure that the greedy, short term optimization does not lead to long term problems, which in our case boils down to two things: not getting into situations where a collision is unavoidable, and making sure that the destination is actually reached. Hence, the main contribution of this paper is to present a trajectory planner with provable safety and task completion properties. Direct methods for trajectory optimization are traditionally based on a priori temporal discretization and collocation methods. In Paper D, the problem of adaptive node distribution is formulated as a constrained optimization problem, which is to be included in the underlying nonlinear mathematical programming problem. The benefits of utilizing the suggested method for  online  trajectory optimization are illustrated by a missile guidance example.In Paper E, the problem of active observer design for an important class of non-uniformly observable systems, namely mobile robotic systems, is considered. The set of feasible configurations and the set of output flow equivalent states are defined. It is shown that the inter-relation between these two sets may serve as the basis for design of active observers. The proposed observer design methodology is illustrated by considering a  unicycle robot model, equipped with a set of range-measuring sensors. Finally, in Paper F, a geometrically intrinsic observer for Euler-Lagrange systems is defined and analyzed. This observer is a generalization of the observer proposed by Aghannan and Rouchon. Their contractivity result is reproduced and complemented  by  a proof  that the region of contraction is infinitely thin. Moreover, assuming a priori bounds on the velocities, convergence of the observer is shown by means of Lyapunov's direct method in the case of configuration manifolds with constant curvature. / QC 20100622 / TAIS, AURES

<p>The main body of this thesis consists of four appended papers. The first two consider different aspects of the trajectory planning problem, while the last two deal with observer design for mobile robotic and Euler-Lagrange systems respectively.</p><p>The first paper addresses the problem of designing a real time, high performance trajectory planner for aerial vehicles. The main contribution is two-fold. Firstly, by augmenting a novel safety maneuver at the end of the planned trajectory, this paper extends previous results by having provable safety properties in a 3D setting. Secondly, assuming initial feasibility, the planning method is shown to have finite time task completion. Moreover, in the second part of the paper, the problem of simultaneous arrival of multiple aerial vehicles is considered. By using a time-scale separation principle, one is able to adopt standard Laplacian control to this consensus problem, which is neither unconstrained, nor first order.</p><p>Direct methods for trajectory optimization are traditionally based on<i> a</i> <i>priori </i>temporal discretization and collocation methods. In the second paper, the problem of adaptive node distribution is formulated as a constrained optimization problem, which is to be included in the underlying nonlinear mathematical programming problem. The benefits of utilizing the suggested method for online trajectory optimization are illustrated by a missile guidance example.</p><p>In the third paper, the problem of active observer design for an important class of non-uniformly observable systems, namely mobile robotics systems, is considered. The set of feasible configurations and the set of output flow equivalent states are defined. It is shown that the inter-relation between these two sets may serve as the basis for design of active observers. The proposed observer design methodology is illustrated by considering a unicycle robot model, equipped with a set of range-measuring sensors.</p><p>Finally, in the fourth paper, a geometrically intrinsic observer for Euler-Lagrange systems is defined and analyzed. This observer is a generalization of the observer recently proposed by Aghannan and Rouchon. Their contractivity result is reproduced and complemented by a proof that the region of contraction is infinitely thin. However, assuming <i>a</i> <i>priori </i>bounds on the velocities, convergence of the observer is shown by means of Lyapunov's direct method in the case of configuration manifolds with constant curvature.</p>

The main body of this thesis consists of four appended papers. The first two consider different aspects of the trajectory planning problem, while the last two deal with observer design for mobile robotic and Euler-Lagrange systems respectively. The first paper addresses the problem of designing a real time, high performance trajectory planner for aerial vehicles. The main contribution is two-fold. Firstly, by augmenting a novel safety maneuver at the end of the planned trajectory, this paper extends previous results by having provable safety properties in a 3D setting. Secondly, assuming initial feasibility, the planning method is shown to have finite time task completion. Moreover, in the second part of the paper, the problem of simultaneous arrival of multiple aerial vehicles is considered. By using a time-scale separation principle, one is able to adopt standard Laplacian control to this consensus problem, which is neither unconstrained, nor first order. Direct methods for trajectory optimization are traditionally based on a priori temporal discretization and collocation methods. In the second paper, the problem of adaptive node distribution is formulated as a constrained optimization problem, which is to be included in the underlying nonlinear mathematical programming problem. The benefits of utilizing the suggested method for online trajectory optimization are illustrated by a missile guidance example. In the third paper, the problem of active observer design for an important class of non-uniformly observable systems, namely mobile robotics systems, is considered. The set of feasible configurations and the set of output flow equivalent states are defined. It is shown that the inter-relation between these two sets may serve as the basis for design of active observers. The proposed observer design methodology is illustrated by considering a unicycle robot model, equipped with a set of range-measuring sensors. Finally, in the fourth paper, a geometrically intrinsic observer for Euler-Lagrange systems is defined and analyzed. This observer is a generalization of the observer recently proposed by Aghannan and Rouchon. Their contractivity result is reproduced and complemented by a proof that the region of contraction is infinitely thin. However, assuming a priori bounds on the velocities, convergence of the observer is shown by means of Lyapunov's direct method in the case of configuration manifolds with constant curvature. / QC 20101108