Cette thèse présente un nouveau cadre pour la conception de correcteurs à gain programmé. Une partie de ce cadre est une fusion novatrice de la théorie des systèmes et de la commande (la métrique ν-gap et sa variante fréquentiellle) et de l'analyse en grappes, technique commune en analyse de données statistiques, apprentissage automatique, fouille de données, etc. La combinaison des deux champs permet de subdiviser le domaine de fonctionnement d'un système non linéaire en secteurs afin de récupérer des informations sur le comportement en boucle fermée avant la conception de la commande. Chaque secteur représente une partie du domaine opérationnel ayant des propriétés de retour similaires, c'est-à-dire que les points de fonctionnement dans un secteur ont des comportements davantage similaires (mesurés par la mesure d'écart ponctuel) les uns des autres que les points de fonctionnement des autres secteurs. La solution de sectorisation est utilisée en vue de réaliser des correcteurs séquencés réglés à partir d'un modèle linéarisé. Par exemple, une distribution optimisée et parcimonieuse des points de synthèse pour les correcteurs LTI est sélectionnée et la distribution des secteurs est exploitée pour le mélange des correcteurs linéaires individuels en un correcteur non-linéaire couvrant l'ensemble du domaine de fonctionnement. L'avantage général de ce cadre est qu'il présente une procédure systématique qui réduit potentiellement le temps, les efforts et donc le coût global d'un projet de développement en réduisant les itérations inutiles au cours du cycle de conception. Le cadre proposé est évalué à partir d’un exemple générique de missile industriel. / This thesis presents a new framework for the design of gain-scheduled controllers. Part of this framework is a novel merging of system & control theory (the ν-gap and pointwise gap metric) and cluster analysis, a common technique in statistical data analysis, machine learning, data mining, etc. The combination of both fields allows for a subdivision of a nonlinear system's operating domain into sectors in order to retrieve information on the feedback behaviour before the actual control design. Each sector represents a part of the operating domain with similar feedback properties, i.e. operating points inside a sector are more similar (as measured by the pointwise gap metric) to each other than to operating points in other sectors. The sectoring solution is used in the proposed framework to support the design of a linearization-based gain-scheduled controller. For example, a reduced and optimized distribution of design points for the LTI controllers is selected and the sectors' distribution is exploited for the blending of the individual linear controllers into an operating domain wide nonlinear controller. The overall advantage of the framework is that it presents a systematic procedure that potentially reduces the overall time, effort, and therefore cost of a development project by preventing unnecessary iterations in the design cycle specifically associated with the control design. The proposed framework is verified at the example of a generic industrial missile benchmark.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND AERO / Bibliography: p. 131-132. / by Louis Howard Knotts. / M.S.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND AERO / Bibliography: leaves B1-B6. / by Michael Roy Mantz. / M.S.

This research study is the result of a 3 years CIFRE PhD thesis between the Airbus design office(Aircraft Control domain) and TéSA laboratory in Toulouse. The main goal is to propose, developand validate a software solution for the detection and diagnosis of a specific type of elevator andrudder vibration, called limit cycle oscillation (LCO), based on existing signals available in flightcontrol computers on board in-series aircraft. LCO is a generic mathematical term defining aninitial condition-independent periodic mode occurring in nonconservative nonlinear systems. Thisstudy focuses on the LCO phenomenon induced by mechanical freeplays in the control surface ofa civil aircraft. The LCO consequences are local structural load augmentation, flight handlingqualities deterioration, actuator operational life reduction, cockpit and cabin comfort deteriorationand maintenance cost augmentation. The state-of-the-art for freeplay induced LCO detection anddiagnosis is based on the pilot sensitivity to vibration and to periodic freeplay check on the controlsurfaces. This study is thought to propose a data-driven solution to help LCO and freeplaydiagnosis. The goal is to improve even more aircraft availability and reduce the maintenance costsby providing to the airlines a condition monitoring signal for LCO and freeplays. For this reason,two algorithmic solutions for vibration and freeplay diagnosis are investigated in this PhD thesis. Areal time detector for LCO diagnosis is first proposed based on the theory of the generalized likeli hood ratio test (GLRT). Some variants and simplifications are also proposed to be compliantwith the industrial constraints. In a second part of this work, a mechanical freeplay detector isintroduced based on the theory of Wiener model identification. Parametric (maximum likelihoodestimator) and non parametric (kernel regression) approaches are investigated, as well as somevariants to well-known nonparametric methods. In particular, the problem of hysteresis cycleestimation (as the output nonlinearity of a Wiener model) is tackled. Moreover, the constrainedand unconstrained problems are studied. A theoretical, numerical (simulator) and experimental(flight data and laboratory) analysis is carried out to investigate the performance of the proposeddetectors and to identify limitations and industrial feasibility. The obtained numerical andexperimental results confirm that the proposed GLR test (and its variants/simplifications) is a very appealing method for LCO diagnostic in terms of performance, robustness and computationalcost. On the other hand, the proposed freeplay diagnostic algorithm is able to detect relativelylarge freeplay levels, but it does not provide consistent results for relatively small freeplay levels. Moreover, specific input types are needed to guarantee repetitive and consistent results. Further studies should be carried out in order to compare the GLRT results with a Bayesian approach and to investigate more deeply the possibilities and limitations of the proposed parametric method for Wiener model identification.

This dissertation addresses the control problem for the general class of control non-affine, non-standard singularly perturbed continuous-time systems. The problem of control for nonlinear multiple time scale systems is addressed here for the first time in a systematic manner. Toward this end, this dissertation develops the theory of feedback passivation for non-affine systems. This is done by generalizing the Kalman-Yakubovich-Popov lemma for non-affine systems. This generalization is used to identify conditions under which non-affine systems can be rendered passive. Asymptotic stabilization for non-affine systems is guaranteed by using these conditions along with well-known passivity-based control methods. Unlike previous non-affine control approaches, the constructive static compensation technique derived here does not make any assumptions regarding the control influence on the nonlinear dynamical model. Along with these control laws, this dissertation presents novel hierarchical control design procedures to address the two major difficulties in control of multiple time scale systems: lack of an explicit small parameter that models the time scale separation and the complexity of constructing the slow manifold. These research issues are addressed by using insights from geometric singular perturbation theory and control laws are designed without making any assumptions regarding the construction of the slow manifold. The control schemes synthesized accomplish asymptotic slow state tracking for multiple time scale systems and simultaneous slow and fast state trajectory tracking for two time scale systems. The control laws are independent of the scalar perturbation parameter and an upper bound for it is determined such that closed-loop system stability is guaranteed. Performance of these methods is validated in simulation for several problems from science and engineering including the continuously stirred tank reactor, magnetic levitation, six degrees-of-freedom F-18/A Hornet model, non-minimum phase helicopter and conventional take-off and landing aircraft models. Results show that the proposed technique applies both to standard and non-standard forms of singularly perturbed systems and provides asymptotic tracking irrespective of the reference trajectory. This dissertation also shows that some benchmark non-minimum phase aerospace control problems can be posed as slow state tracking for multiple time scale systems and techniques developed here provide an alternate method for exact output tracking.

The helicopter is a Multiple-Input Multiple-Output (MIMO) system with highly coupled characteristics, which increases the complexity of the system dynamics. In addition, the system dynamics of the helicopter are unstable, referring to its tendency to deviate from an equilibrium when disturbed. Despite the complexity in its modelling and control, the benefit of using a helicopter for unmanned, autonomous applications can be tremendous. One particular application that motivates this research is the use of an unmanned small-scale helicopter in an autonomous survey mission over an area struck by disaster, such as an earthquake. The work presented in this thesis provides a framework for utilizing a platform system for research and development of small-scale helicopter systems. A platform system enables testing and analysis to be performed indoor in a controlled environment. This can provide a more convenient mean for helicopter research since the system is not affected by environmental elements, such as wind, rain or snow condition. However, the presence of the platform linkages poses challenges for analysis and controller design as it alters the helicopter system flight dynamics. Through a six degree-of-freedom (6 DOF) platform model derived in this research, the criteria for matching the trim conditions between the platform system and a stand alone helicopter have been identified. With the matched trim conditions, linearization is applied to perform analysis on the effects that the platform has on the system dynamics. The results of the analysis provide insights into both the limitations and benefits of utilizing the platform system for helicopter research. Finally, a Virtual Joint Control scheme is proposed as an unified control strategy for both the platform and the stand alone helicopter systems. Having a consistent control scheme between the two systems allows for comparisons between simulation and experimental results for the two systems to be made more readily. Furthermore, the Virtual Joint Control scheme represents a novel flight control strategy for stand alone helicopter systems.

In this thesis we study a number of nonlinear control problems motivated by their appearance in flight control. The results are presented in a general framework and can also be applied to other areas. The two main topics are backstepping and control allocation. Backstepping is a nonlinear control design method that provides an alternative to feedback linearization. Here, backstepping is used to derive robust linear control laws for two nonlinear systems, related to angle of attack control and flight path angle control, respectively. The resulting control laws require less modeling information than corresponding designs based on feedback linearization, and achieve global stability in cases where feedback linearization can only be performed locally. Further, a method for backstepping control of a rigid body is developed, based on a vector description of the dynamics. We also discuss how to augment an existing nonlinear controller to suppress constant input disturbances. Two methods, based on adaptive backstepping and nonlinear observer design, are proposed. Control allocation deals with actuator utilization for overactuated systems. In this thesis we pose the control allocation problem as a constrained least squares problem to account for actuator position and rate constraints. Efficient solvers based on active set methods are developed with similar complexity to existing, approximate, pseudoinverse methods. A method for dynamic control allocation is also proposed which enables a frequency dependent control distribution among the actuators to be designed. Further, the relationship between control allocation and linear quadratic control is investigated. It is shown that under certain circumstances, the two techniques give the same freedom in distributing the control effort among the actuators. An advantage of control allocation, however, is that since the actuator constraints are considered, the control capabilities of the actuator suite can be fully exploited.

The helicopter is a Multiple-Input Multiple-Output (MIMO) system with highly coupled characteristics, which increases the complexity of the system dynamics. In addition, the system dynamics of the helicopter are unstable, referring to its tendency to deviate from an equilibrium when disturbed. Despite the complexity in its modelling and control, the benefit of using a helicopter for unmanned, autonomous applications can be tremendous. One particular application that motivates this research is the use of an unmanned small-scale helicopter in an autonomous survey mission over an area struck by disaster, such as an earthquake. The work presented in this thesis provides a framework for utilizing a platform system for research and development of small-scale helicopter systems. A platform system enables testing and analysis to be performed indoor in a controlled environment. This can provide a more convenient mean for helicopter research since the system is not affected by environmental elements, such as wind, rain or snow condition. However, the presence of the platform linkages poses challenges for analysis and controller design as it alters the helicopter system flight dynamics. Through a six degree-of-freedom (6 DOF) platform model derived in this research, the criteria for matching the trim conditions between the platform system and a stand alone helicopter have been identified. With the matched trim conditions, linearization is applied to perform analysis on the effects that the platform has on the system dynamics. The results of the analysis provide insights into both the limitations and benefits of utilizing the platform system for helicopter research. Finally, a Virtual Joint Control scheme is proposed as an unified control strategy for both the platform and the stand alone helicopter systems. Having a consistent control scheme between the two systems allows for comparisons between simulation and experimental results for the two systems to be made more readily. Furthermore, the Virtual Joint Control scheme represents a novel flight control strategy for stand alone helicopter systems.

Utilizing the universal approximation property of neural networks, we develop several novel approaches to neural network-based adaptive output feedback control of nonlinear systems, and illustrate these approaches for several flight control applications. In particular, we address the problem of non-affine systems and eliminate the fixed point assumption present in earlier work. All of the stability proofs are carried out in a form that eliminates an algebraic loop in the neural network implementation. An approximate input/output feedback linearizing controller is augmented with a neural network using input/output sequences of the uncertain system. These approaches permit adaptation to both parametric uncertainty and unmodeled dynamics. All physical systems also have control position and rate limits, which may either deteriorate performance or cause instability for a sufficiently high control bandwidth. Here we apply a method for protecting an adaptive process from the effects of input saturation and time delays, known as ``pseudo control hedging". This method was originally developed for the state feedback case, and we provide a stability analysis that extends its domain of applicability to the case of output feedback. The approach is illustrated by the design of a pitch-attitude flight control system for a linearized model of an R-50 experimental helicopter, and by the design of a pitch-rate control system for a 58-state model of a flexible aircraft consisting of rigid body dynamics coupled with actuator and flexible modes. A new approach to augmentation of an existing linear controller is introduced. It is especially useful when there is limited information concerning the plant model, and the existing controller. The approach is applied to the design of an adaptive autopilot for a guided munition. Design of a neural network adaptive control that ensures asymptotically stable tracking performance is also addressed.

The development and testing for a UAV testbed to be used in academic research and undergraduate education is proposed in this thesis. Analysis on commercial off-the-shelf UAV systems and autopilots lead to the development of a custom, open-architecture and modular UAV testbed. The main focus is to support research in UAV control field and education of the undergraduate students. The integration and use of commercial-off-the-shelf avionics and air vehicle are described in detail. System performance is examined both in flight and on the ground. Results of the system tests show that the developed system is a functional UAV testbed to be used in research of different flight control algorithms.