Tethered satellites systems have been studied for several decades from which the concept of tether-connected space manipulators developed. The advantages of such robots derive from the extensive reach that can be provided by the tether. This would benefit applications such as satellite repair and space debris removal. / This thesis first examines the dynamics of a tethered space manipulator as its end-effector traces a prescribed trajectory. The platform, to which the robot arm is attached, is postulated to be much larger than the other components of the system and is specified to follow a circular orbit around Earth. The tether is modelled as a straight rigid rod and the motion of the system is assumed to be restricted to the orbital plane. The tether tension is monitored to ensure that the tether is not subjected to compressive forces. / The system is also studied under torque restrictions; namely, the torques acting on the tether attachment points are set to zero. The possible trajectories traced by the end-effector of this constrained system are investigated. In addition, a method is outlined to determine a feasible end-effector path between two given points while satisfying the zero-torque constraint on the tether.

A structural model in the frequency domain is developed to obtain aeroelastic solutions of a NACA64A010 in the transonic regime. Results from a fully nonlinear frequency domain flow solver and the structural model are compared to a state of the art time accurate flutter code. The prohibitive cost of using time accurate methods to solve problems in transonic aeroelasticity is introduced and then followed by an analysis of using a Non-Linear Frequency Domain (NLFD) approach. The computational cost and accuracy associated with using the NLFD method are compared with already established time accurate methods. In this work, an accurate representation of the transonic flutter boundary is obtained and compared to pre-existing numerical results, further substantiating the NLFD method's ability to account for strong nonlinearities in the flow field. The flight conditions considered herein are of a swept wing in the transonic regime, a test case extensively studied by Isogai. This research has further validated the NLFD method as an accurate and efficient alternative for computing aeroelastic solutions in the transonic regime.

Flutter analysis plays an important role in the design and development of aircraft wings because of the information it provides regarding the flight envelope of the aircraft. With the coupling of the flow and structural solver, the flutter boundary of wings can be evaluated in the time domain. This study: First, computes the aeroelastic response for a typical sweptback wing section model by coupling a flow solver and a two degree of freedom structural equation of motion solver to predict the flutter boundary of an airfoil at different Mach numbers. The results agree well with previous numerical results, and the transonic-dip phenomenon can be observed. Second, a new coupling approach is introduced to conservatively transfer the load and displacement between the flow solver and the structural solver for 3-D flow. By coupling the flow solver and a low fidelity finite element structural model, the flutter point of AGARD wing 445.6 at Mach number 0.499 is computed. The flutter point agrees well with experimental results and previous numerical results.

The objective of the present study was to investigate the mechanisms of sound generation by subsonic jets. Large eddy simulations were performed along with bandpass filtering of the flow and sound in order to gain further insight into the pole of coherent structures in subsonic jet noise generation.A sixth-order compact scheme was used for spatial discretization of the fully compressible Navier-Stokes equations. Time integration was performed through the use of the standard fourth-order, explicit Runge-Kutta scheme. An implicit low dispersion, low dissipation Runge-Kutta (ILDDRK) method was developed and implemented for simulations involving sources of stiffness such as flows near solid boundaries, or combustion. A surface integral acoustic analogy formulation, called Formulation 1C, was developed for farfield sound pressure calculations. Formulation 1C was derived based on the convective wave equation in order to take into account the presence of a mean flow. The formulation was derived to be easy to implement as a numerical post-processing tool for CFD codes.Sound radiation from an unheated, Mach 0.9 jet at Reynolds number 400, 000 was considered. The effect of mesh size on the accuracy of the nearfield flow and farfield sound results was studied. It was observed that insufficient grid resolution in the shear layer results in unphysical laminar vortex pairing, and increased sound pressure levels inthe farfield. Careful examination of the bandpass filtered pressure field suggested that there are two mechanisms of sound radiation in unheated subsonic jets that can occur in all scales of turbulence. The first mechanism is the stretching and the distortion of coherent vortical structures, especially close to the termination of the potential core. As eddies are bent or stretched, a portion of their kinetic energy is radiated. This mechanism is quadrupolar in nature, and is responsible for strong sound radiation at aft angles. The second sound generation mechanism appears to be associated with the transverse vibration of the shear-layer interface within the ambient quiescent flow, and has dipolar characteristics. This mechanism is believed to be responsible for sound radiation along the sideline directions.Jet noise suppression through the use of microjets was studied. The microjet injection induced secondary instabilities in the shear layer which triggered the transition to turbulence, and suppressed laminar vortex pairing. This in turn resultedin a reduction of OASPL at almost all observer locations. In all cases, the bandpass filtering of the nearfield flow and the associated sound provides revealing details of the sound radiation process. The results suggest that circumferential modes are significant and need to be included in future wavepacket models for jet noise prediction.Numerical simulations of sound radiation from nonpremixed flames were also performed. The simulations featured the solution of the fully compressible Navier-Stokes equations. Therefore, sound generation and radiation were directly captured in the simulations. A thickened flamelet model was proposed for nonpremixed flames. The model yields artificially thickened flames which can be better resolved on the computational grid, while retaining the physically currect values of the total heat released into the flow. Combustion noise has monopolar characteristics for low frequencies. For high frequencies, the sound field is no longer omni-directional. Major sources of sound appear to be located in the jet shear layer within one potential core length from the jet nozzle. / L'objectif de cette étude est d'obtenir la meilleure compréhension des mécanismesde géneration de bruit par des jet subsoniques. Cette étude est basée sur simulations aux grandes échelles de jets réactifs et sans réactifs. Des calculs numériques employant des schéme compacts de sixiéme ordre. L'integration temporelle fut éxéciteé à l'aide de schéme Runge-Kutta de de quatrième ordre. Des schéme à faible dispersion et dissipation numérique. Un formulation intégrale basée sur les analogies acoustiques fut développées pour la prédiction du champ acoustique lointain pour les sources et observateure en mouvement dans un fluide avec vitesse uniforme. La formulation fut implémentée à l'aide d'algorithmes facilitant l'implémentation pour le traitement de données d'écoulement en haute performance utilisant des outils de simiulation à grande échelle. Les champs sonore produit par un jet turbulent non-réactif avec nombre de Mach de 0.9, et un nombre de Reynolds ReD = 400, 000 fut étudié. L'effect de la taille du maillage sur la précision de l'écoulement en champs proche et e champs sonore loin de source fut analysé. La sous-résolution de la couche de cisaillement à la sortie du jet méne à l'apparition de structures cohérentes et forte radiation qui ne sort pas physiquement réalistes. Deux mécanismes principaux de génération sonore par jets subsoniques furent identifiés. Le premier mécanisme est l'étirement et la distorsion de structures tourbillonnaires cohérentes, en particulier prés de la fin du coere potentiel. Ce mécanisme est quadripolaire, et émet principalement vers l'arriére du jet dans la direction de l'écoulement. Le seconde mécanisme semble être constitué de vibration transversale de la couche de cisaillement en réponse à la présemce de structures cohérentes dans la jet. Semblable à la radiation d'une plaque à bonds finis, la contribution de ce méchanisme est dipolaire et domine la champs sonore dans la direction transversale, perpendiculaire au jet. L'utilisation de plusieurs microjet fut investiguée pour la réduction du bruit. L'injection à l'aide de microjets précipite la transition à la turbulence, favorisent le mélange et la destrcutction de structures cohérentes de grande échelle. Un filtrage en bandes de étroites fut effectué. Ce traitement des données numériquepermet de visualiser les relations complexes entre l'écoulement et les onds sonores émises. Les résultats démontrent l'importance de modes circumférenciele, ce qui a des conséquenecs pour les modiles dits de paquets d'onde pour la preédiction du bruit du jet. Des simulation numériques d'écoulement et champs sonore d'une flame sans prémélange furent aussi éxécutées. Les simulations incluent encore une fois l'écoulement et le champ sonore associé, obtenus directement des équations de Navier Stokes compressibles. Un modèle flammelette épaissie fut proposé que donne flammes épaissies artificiellement qui peuvent être mieux résolus sur le maillage. Le bruit de combustion a des caractéristiques monopolaires aux basses fréquences. Principales sources de bruit semblent être situé dans la couche de cisaillement.

In the past decades, autonomous on-orbit servicing has become a priority for the space industry. One important application is refueling and servicing of satellites. This type of operation requires a system capable of autonomously capturing the client satellite, which is a free-floating object. In this work, a robotic manipulator is used for the interception task. The latter is divided in two phases : the approach and the capture. In the first phase, the pose of the client is estimated and its future trajectory predicted from visual data through a Kalman filter. This information is combined with the robot motion time from a trajectory generator in an iterative procedure to determine the time-optimal interception location. The trajectory of the manipulator is generated from a concatenation of quintics defining the acceleration profile. The robot desired configuration for capture is calculated using the concept of redundancy parameter. In the second phase of the interception, a close-range tracking algorithm is used for fine alignment and capture. The development of the complete interception scheme, simulation and experimental results are presented in this thesis, the latter obtained with the robot-airship facility in the Aerospace Mechatronics Laboratory. / Au cours des dernières décennies, l'utilisation de systèmes autonomes afin d'accomplir des tâches en orbite a reçu de plus en plus d'attention de la part de l'industrie aerospaciale. Le ravitaillement, la mise à jour et la réparation de satellites en orbite représente deux applications intéressantes de cette approche. De telles opérations requièrent un system capable de capturer un satellite défectueux (le client) de façon autonome. Dans le cas présent, un bras manipulateur robotisé est utilisé afin d'intercepter le client. L'interception se divise en deux parties distinctes : l'approche et la capture. La première phase consiste à estimer la pose et la trajectoire future du client à l'aide des données provenant d'un système de vision et d'un filtre de Kalman. Le lieu optimal où l'interception aura lieu est ensuite choisie par iteration en tenant compte des données du filtre et de celles du générateur de trajectoire du robot. Cet dernière est créée par la concaténation de polynômes du cinquième degrée définissant la courbe d'accélération des joints. Lors de la seconde phase, un algorithme de pistage est utilisé afin d'obtenir un alignment plus précis et de capturer le client. Cette thèse présente les détails de la méthode d'interception et expose les résultats de simulations numériques et d'expériences. Les résultats expérimentaux ont été obtenues à l'aide d'un bras manipulateur et d'un ballon à l'hélium imitant la trajectoire d'un satellite dans l'espace.

The aerodynamic problems, involving the solution of the Euler equations of motion, are presently almost exclusively solved using an Euler formulation. This requires the generation of a spatial grid over which the problem is discretized. The grid generation process adds to the complexity to the problem being solved, especially in the case of complex body-shapes with flow discontinuities, such as shock waves and sliplines. ALso, the Euler formulation is difficult to use for the aerodynamic problems of unspecified geometry. / A Lagrangian formulation permits to avoid the complexity of a grid generation while making it possible to obtain very accurate results. This Lagrangian formulation use the stream function and Lagrangian distance to represent the flow instead of the Cartesian coordinates. Thus, the streamlines become coordinate lines in the Lagrangian formulation, which can easily represent complex body-shapes and sliplines, and is also more suitable to solve complex problems involving bodies of unspecified geometry. / This method was tested and validated against several test problems of specified geometry, including the supersonic flows with shock waves in a duct with a circular arc bump and past airfoils, as well as the flow in a nozzle. The solutions obtained with this Lagrangian method were found to be very accurate, displaying a high at a computational efficiency (actually providing second-order accuracy at a computational load of a first order solution). Then, this Lagrangian method has been used to solve several aerodynamic problems with geometrically-unspecified body-shape, such as (i) the determination of the geometry of a bump corresponding to a specified pressure distribution in supersonic flow, and (ii) the design of the geometry of a supersonic nozzle, based on the reflection-suppression condition, for a specified uniform flow at exit.

The equations of motion of a tethered satellite system are highly nonlinear and should possess many interesting related features; yet its nonlinear dynamics has never been thoroughly investigated in previous works. This thesis analyzes the nonlinear dynamics of two-body tethered satellite systems using numerical tools of analysis such as phase plane plots, power spectral densities (PSD's), Poincare sections and first Lyapunov exponents, as well as approximate analytical methods including the method of Melnikov. Motion in the stationkeeping phase wherein the tethered system is just a gravity gradient pendulum is studied, first considering pitch motion only, and then considering the coupled pitch and roll motions. The deployment/retrieval phases are studied next. For a circular orbit, pitch stability is examined for varying exponential length rates; for the unstable cases, it is compared to an equivalent uniform length rate scheme, which showed better stability behaviour. (Abstract shortened by UMI.)

In this thesis, the problem of attitude dynamics and maneuvering of flexible, multibody space systems is studied. A formulation for deriving the equations of motion of these systems, based on Kane's method, is presented. In this formulation the concepts of constrained motion and the effect of nonlinear coupling between rigid-body motion and the elastic vibrations are examined in depth. / Dynamics of constrained systems is studied with the main objective of deriving the complete, minimum-dimension set of equations of motion. / The effect of rigid-body motion on the dynamic behavior of flexible systems, known as dynamic or geometric stiffening effect, is examined in detail. / A formulation is developed and implemented in a symbolic computer code, FLXSIM, for deriving the analytical form of the equations of motion. / Application of artificial constrained motion to devise open-loop control laws for tracking problems is proposed. / A perturbation technique in conjunction with a phase-plane based optimal control analysis is proposed for near-minimum-time maneuvering of flexible multibody systems moving along a prescribed trajectory. (Abstract shortened by UMI.)

The design and mechanical performance of a helicopter horizontal stabilizer slat made by resin transfer moulding (RTM) can be evaluated with finite element analysis (FEA). To verify the validity of the assumptions used in the modelling of the slat structure, static mechanical tests were performed on prototype slats which were half of the full-size length. The slat complex boundary conditions were simplified to make static mechanical testing possible. Two fixtures were designed and built to introduce simplified loads in specimens with two different bracket configurations: a full and a half bracket. A finite element (FE) model of the specimens was made with shell elements and the finite element solution was compared with the experimental results. In most cases, comparison between the finite element analysis solution and experimental results showed good agreement in terms of structure stiffness, strength, strain and damage location. It is believed that out-of-plane stresses should be considered to improve the finite element solution accuracy. / Le design et la performance mécanique d’un bec de bord d’attaque de stabilisateur horizontal d’hélicoptère fabriqué par moulage par injection sur renfort (« resin transfer moulding ») peuvent être évalués par des analyses par éléments finis. Pour vérifier la validité des hypothèses utilisées dans la modélisation du bec de bord d’attaque, des essais mécaniques ont été effectués sur des prototypes demi-longueurs. Les conditions frontières complexes ont été simplifiées pour rendre les essais mécaniques possibles. Deux gabarits ont été conçus et construits pour introduire les chargements simplifiés dans les pièces d’essai possédant deux configurations de support: un support complet et un demi-support. Un modèle par éléments finis des pièces d’essai a été réalisé avec des éléments de type membrane et la solution est comparée avec les résultats expérimentaux. Dans la plupart des cas, la comparaison entre la solution par éléments finis et les résultats expérimentaux coïncide concernant la rigidité de la structure, la résistance, l’allongement et la localisation de l’endommagement. Les contraintes hors du plan devraient être considérées pour améliorer la précision de la solution par éléments finis.

The thesis focuses on efficient solutions of non-cooperative pursuit-evasion games with imperfect information on the state of the system. This problem is important in the context of interception of future maneuverable ballistic missiles. However, the theoretical developments are expected to find application to a broad class of hybrid control and estimation problems in industry. The validity of the results is nevertheless confirmed using a benchmark problem in the area of terminal guidance. A specific interception scenario between an incoming target with no information and a single interceptor missile with noisy measurements is analyzed in the form of a linear hybrid system subject to additive abrupt changes. / The general research is aimed to achieve improved homing accuracy by integrating ideas from detection theory, state estimation theory and guidance. The results achieved can be summarized as follows. (i) Two novel maneuver detectors are developed to diagnose abrupt changes in a class of hybrid systems (detection and isolation of evasive maneuvers): a new implementation of the GLR detector and the novel adaptive- H0 GLR detector. (ii) Two novel state estimators for target tracking are derived using the novel maneuver detectors. The state estimators employ parameterized family of functions to described possible evasive maneuvers. (iii) A novel adaptive Bayesian multiple model predictor of the ballistic miss is developed which employs semi-Markov models and ideas from detection theory. (iv) A novel integrated estimation and guidance scheme that significantly improves the homing accuracy is also presented. The integrated scheme employs banks of estimators and guidance laws, a maneuver detector, and an on-line governor; the scheme is adaptive with respect to the uncertainty affecting the probability density function of the filtered state. (v) A novel discretization technique for the family of continuous-time, game theoretic, bang-bang guidance laws is introduced. The performance of the novel algorithms is assessed for the scenario of a pursuit-evasion engagement between a randomly maneuvering ballistic missile and an interceptor. Extensive Monte Carlo simulations are employed to evaluate the main statistical properties of the algorithms. (Abstract shortened by UMI.)