Дисертації з теми "Systèmes adaptatifs (technologie)"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Systèmes adaptatifs (technologie)".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Arrouf, Mhamed. "Optimisation multicritère appliquée aux systèmes mécaniques adaptatifs." Besançon, 2004. http://www.theses.fr/2004BESA2007.
Повний текст джерелаThe aim of this work is to extend multi-criterion optimisation methods to active structure design, and particularly to the placement of sensors and actuators on a mechanical structure. Adaptive mechanical systems use LQG regulators (Linear Quadratic Gaussien). The main matrices are determined by minimisation of a quadratic criterion and under consideration of some constraints that are used during the iterative process (ex: the use of actuators at the maximum of their possibilities leads to iteratively modify the specific ratio of weight factors). We propose an original strategy for the optimisation of the regulator gain matrix by considering constraints on the actuator control and outputs. Since the problem is an optimisation under constraints, UZAWA algorithm was used to the search of saddle-point. The choice and placement of actuators and sensors are of course made according to the fixed objectives, but it is also leaded by taking into account simple ideas given by an expert system knowledge. Fuzzy logic approach was used to optimise actuators placement. This approach allows the consideration of knowledge given by expert as well as the easy introduction of additional expert rules. Finally, ways to the first sensor placement on the active structure are shown using energetic standards for perturbability and observability. This work was applied to the study of an aluminium panel
Le, Bras Florent. "Contributions à l'asservissement visuel de véhicules aériens autonomes." Nice, 2010. http://www.theses.fr/2010NICE4110.
Повний текст джерелаThe visual servo control technology is based on measurements form a camera. This technique appeared in the late 80’s to control mobile robots or manipulators. Recently, visual servoing is applied to flying vehicles, for which this type of control has a strong operational potential. In this thesis, we focus on the design of visual servoing for two types of flying vectors : the VTOL minidrones and fixed-wing aircrafts. We try wherever possible to specify the problem in terms of regulation in the picture by taking into account the limitations of the sensors used. Thus we consider only the onboard sensors autonomous, and whose quality is consistent with a the considered visual servoing tasks. For this, we propose visual servoing laws adapted to the considered systems intheir control structures and their measurement systems. For VTOL minidrones, a series of control laws is proposed to stabilize the vehicle with respect to a relativity flat target which some landmarks are detected in real time. These commands have the distinction of not requiring measurements of the speed of translation. This characteristic is especially interesting, since this variable is particularly difficult to measure without GPS. One of the algorithms has been, validated in fight ion the platform. HoverEye developed by Bertin Technologies, it was, to our knowledge the first image based visual servoing, conducted indoors on a minidrones. For fixed-wing aircrafts, a next autonomous landing technique based on the detection of runway’s edges is proposed. This technique is image based and takes into account the specificities of aircraft flight dynamics. Moreover a specific estimate of cross-wind is proposed. Validation of this approach is performed on a complex simulator which includes images processing. Finally, we sought new applications in image based visual servoing for fixed-wing aircrafts. A control law for precise stabilization on observation orbit has been synthesized. It is based on the detection of a single landmark and translational optical flow. Simulations illustrate the performance of that algorithm
Feki, Mohamed Ali. "An ontology based context aware modelling and reasoning to enhance human environment interaction." Evry, Institut national des télécommunications, 2007. http://www.theses.fr/2007TELE0008.
Повний текст джерелаThe recent convergence of ubiquitous computing and context-aware computing has seen a considerable rise in interest that exploit aspects of the contextual environment to enhance implicit user interaction, offer services, present information, tailor application behavior or trigger adaptation. However, as a result of the lack of generic mechanisms for supporting context-awareness, context-aware applications remain very difficult to build and developers must deal with a wide range of issues related to representing, sensing, aggregating, storing, querying and reasoning of context. In order to remedy this situation, there is a need for better understanding of the design process associated with context-aware applications, architectural support for the entire context processing flow, and improved programming abstractions that ease the prototyping of applications. This research in context-aware computing has focused on the architectural support for context-aware application development. This dissertation presents a set of requirements for context-aware applications, based on wich we designed and implemented our architectural support, including an ontology-based context model, a context-aware architecture (namely Context Aware Explorer) and a set of programming abstractions. This research reported here is investigating : how context can be acquired, ditributed, and used and how it changes human computer interaction in Ubiquitous Computing. The Context Aware Explorer includes common steps required to build context applications (acquisition, interpretation, presentation, reasoning and application), thus it ensures the adaptation situated at the level of User Environment Interaction. The case study, Assistive Environment, validates our work, and illustrates, in concrete form, the process and issues involved in the design of context-aware software. Finally, the research presents an overview of how Ubiquitous Computing systems can be evaluated. Different techniques are assessed, and the concept of probing users and developers with prototypes is presented
Rebaud, Bettina. "Evolution vers des architectures de systèmes intégrés auto-adaptatives et tolérantes aux variations technologiques et environnementales." Montpellier 2, 2009. http://www.theses.fr/2009MON20228.
Повний текст джерелаWith the « More Moore » and low power trends, optimizing or only well predicting the final performances of digital circuits become more and more difficult. Indeed, variability and hardness to accurately model transistor behavior impede the dimension scaling benefits. Current design methods generally use guard margins to prevent from the incertitude generated by these limits and to guarantee functional yield. But as we go in the nanometer era, the margin use is not efficient anymore, because of an increasing over-design, limiting optimizations and decreasing yield. Very different numerous solutions exist in order to overcome these troubles, following two main goals which are dissimilar but also complementary: increasing the robustness to uncertainty during the design levels, notably thanks to better performance analysis; and adapting the final circuit to its real process corner and to its environment. In this work, we considered these two directions: the first one thanks to (1) an implementation of a specific methodology called SSTA (Statistical Static Timing Analysis), allowing to perform statistical analysis on timing performances; this methodology allows us to accurately observe process variation effects on delays; and the second one in proposing (2) an efficient diagnostic system based on the in situ critical path monitoring concept and allowing all adaptive solution implementations. The latter answer to variability issue allows to know the real characteristics of the circuit, to crop design margins, to improve power consumption and frequency performances, and even to increase yield
Belaggoun, Amel. "Adaptability and reconfiguration of automotive embedded systems." Electronic Thesis or Diss., Paris 6, 2017. http://www.theses.fr/2017PA066252.
Повний текст джерелаModern vehicles have become increasingly computerized to satisfy the more strict safety requirements and to provide better driving experiences. Therefore, the number of electronic control units (ECUs) in modern vehicles has continuously increased in the last few decades. In addition, advanced applications put higher computational demand on ECUs and have both hard and soft timing constraints, hence a unified approach handling both constraints is required. Moreover, economic pressures and multi-core architectures are driving the integration of several levels of safety-criticality onto the same platform. Such applications have been traditionally designed using static approaches; however, static approaches are no longer feasible in highly dynamic environments due to increasing complexity and tight cost constraints, and more flexible solutions are required. This means that, to cope with dynamic environments, an automotive system must be adaptive; that is, it must be able to adapt its structure and/or behaviour at runtime in response to frequent changes in its environment. These new requirements cannot be faced by the current state-of-the-art approaches of automotive software systems. Instead, a new design of the overall Electric/Electronic (E/E) architecture of a vehicle needs to be developed. Recently, the automotive industry agreed upon changing the current AUTOSAR platform to the “AUTOSAR Adaptive Platform”. This platform is being developed by the AUTOSAR consortium as an additional product to the current AUTOSAR classic platform. This is an ongoing feasibility study based on the POSIX operating system and uses service-oriented communication to integrate applications into the system at any desired time. The main idea of this thesis is to develop novel architecture concepts based on adaptation to address the needs of a new E/E architecture for Fully Electric Vehicles (FEVs) regarding safety, reliability and cost-efficiency, and integrate these in AUTOSAR. We define the ASLA (Adaptive System Level in AUTOSAR) architecture, which is a framework that provides an adaptive solution for AUTOSAR. ASLA incorporates tasks-level reconfiguration features such as addition, deletion and migration of tasks in AUTOSAR. The main difference between ASLA and the Adaptive AUTOSAR platform is that ASLA enables the allocation of mixed critical functions on the same ECU as well as time-bound adaptations while adaptive AUTOSAR separates critical, hard real-time functions (running on the classic platform) from non-critical/soft-real-time functions (running on the adaptive platform). To assess the validity of our proposed architecture, we provide an early prototype implementation of ASLA and evaluate its performance through experiments
Saidi, Taofik. "Architectures matérielles pour la technologie W-CDMA étendue aux systèmes multi-antennes." Rennes 1, 2008. ftp://ftp.irisa.fr/techreports/theses/2008/tsaidi.pdf.
Повний текст джерелаOver the last ten years, multi-antenna systems (MIMO) for mobile and fixed wireless communications have revolutionized the possibilities offered for numerous telecommunication application domains. Using multiple antennas at both ends of the link considerably increases the capacity of wireless systems. However, the digital algorithms needed to realize these systems are complex and constitute a challenge in order to define powerful hardware architectures. The goal of the present work centers specifically on the optimal definition of architectural solutions to combat this problem in a CDMA context. The first aspect of this work is a deepened study of space-time algorithms and design methods with regards to an efficient hardware implementation. Numerous detection schemes are proposed in the litterature and are applicable following three criteria : quality of service, binary throughput and algorithmic complexity. The latter constitutes a strong limitation for a low cost implementation of mobiles including these applications. Thus, it is necessary to use powerful tools to simulate, evaluate and rapidly prototype these new systems which constitute likely candidates for fourth generation telecommunication systems. The second aspect is the realization of an uncoded multi-antenna transceiver, integrating CDMA in a wideband channel case. A WCDMA mono-antenna system, generalized to any antenna array, has been integrated and simulated on the Lyrtech rapid prototyping platform. The developped architecture integrates the main baseband processing modules, such as Nyquist filtering, multiple path detection and detection itself. The MIMO-WCDMA prototype is characterized by its flexibility with regards to the number of inputs, format of samples, characteristics of the wireless channel and the targeted technology (ASIC, FPGA). The third aspect is more prospective since it introduces new methods to reduce the hardware cost of multi-antenna systems. The principle of dynamic allocation of the fixed point format is presented with the goal of adapting the data encoding according to the wireless channel's characteristics and consequently minimizing the circuit's complexity. Also, the concept of adaptive architectures is proposed to reduce the power consumption in an embedded system according to application context
Rouissi, Firas. "Contribution à l'analyse des propriétés d'observabilité et de diagnosticabilité des systèmes complexes. Application à une éolienne." Rouen, 2013. http://www.theses.fr/2013ROUES032.
Повний текст джерелаAtig, Asma. "Sur la commande neuronale adaptative des systèmes non linéraires." Le Havre, 2012. http://www.theses.fr/2012LEHA0004.
Повний текст джерелаThis work deals with the control of non linear square MIMO systems with unknown or uncertain dynamics. The retained adaptative neural control scheme consists of an emulator and a controller. The emulator and controller parameters are decoupled and adjusted in real time using the Real Time Recurrent Learning algorithm. New adptation approaches for the controller parameters, based on the Lyapunov stability study, are proposed to ensure a stable behavior of disturbed nonlinear systems. Minimization of multi-objective criteria is also considered to adjust the controller parameter. Performances of the proposed approaches were evaluated by numerical simulations and by real time applications on disturbed nonlinear systems
Pley, Julien. "Protocoles d'accord pour la gestion d'une grille de calcul dynamique." Rennes 1, 2007. ftp://ftp.irisa.fr/techreports/theses/2007/pley.pdf.
Повний текст джерелаWe present a middleware for dynamic grids where the federated resources are provided by different institutions. Within a domain, the resources interact in a synchronous manner. Interactions between resources belonging to different domains are asynchronous. Every machine or domain can join or leave the grid at any time; due to failures, or on purpose. We propose a fault-tolerant solution which takes advantage of this hierarchical structure to solve the grid membership problem and the load-balanced task allocation problem. Each service is the composition of a synchronous protocol and an asynchronous agreement protocol which is always a variation of the fundamental Consensus problem. We define the "insensitivity to erroneous suspicions"; a new metric to compare different Consensus protocols based on Diamond S failure detectors
Dorie, Laurent. "Modélisation et évaluation de performances en vue de la conception conjointe des systèmes reconfigurables : application à la radio logicielle." Nantes, 2007. http://www.theses.fr/2007NANT2107.
Повний текст джерелаThe fast evolution of embedded system context leads to more and more complexity into electronic products that can support many ways of working and different standards. In these systems, the reconfiguration is a solution to face such evolution and also respect embedded constraints. This property points out that a system is able to modify its behaviour. Such property concerns just as well the application development as the technology design. New approaches and tools are needed to take into account this reconfiguration property. Thus, the goal of this thesis is to provide high abstraction level models in order to improve the co-design of reconfigurable systems. The first part of this thesis interested in reconfiguration mechanisms of radiocommunication systems. It led to the definition of modelling in order to describe the reconfigurable mechanisms of radio communication application. The second part of this thesis focused on the reconfigurable architectures. It led to a modelling able to describe the reconfigurable impact of heterogeneous multi-processor platforms on system behaviour and performances. The interest of these modelling is illustrated by a study which deals with a typical case of Software Radio
Darouich, Mehdi. "Reefs : une architecture reconfigurable pour la stéréovision embarquée en contexte temps-réel." Rennes 1, 2010. http://www.theses.fr/2010REN1S151.
Повний текст джерелаStereovision allows the extraction of depth information from several images taken from different points of view. In computer vision, stereovision is used to evaluate directly and accurately the distance of objects. In Advanced Driver Assistance Systems (ADAS), number of applications needs an accurate knowledge of the surrounding and can thus benefit from 3D information provided by stereovision. Involved tasks are done in real-time and require a high level of performance that can be provided by hardware accelerators. Moreover, as people safety is affected, the reliability of results is critical. As a result, the hardware solution has to be flexible enough to allow this adaptation. Finally, as the embedded context is considered, the silicon area of the chosen hardware solution must be limited. The purpose of this thesis is to design a processing architecture for stereovision that provides a performance level answering ADAS requirements and a level of flexibility high enough to generate depth maps adapted to various applications. A heterogeneous reconfigurable architecture, named REEFS (Reconfigurable Embedded Engine for Flexible Stereovision), is designed and scaled to answer ADAS requirements and to provide the best trade-off between flexibility, performance and silicon area
Azar, Céline. "On the design of a distributed adaptive manycore architecture for embedded systems." Lorient, 2012. http://www.theses.fr/2012LORIS268.
Повний текст джерелаChip design challenges emerged lately at many levels: the increase of the number of cores at the hardware stage, the complexity of the parallel programming models at the software level, and the dynamic requirements of current applications. Facing this evolution, the PhD thesis aims to design a distributed adaptive manycore architecture, named CEDAR (Configurable Embedded Distributed ARchitecture), which main assets are scalability, flexibility and simplicity. The CEDAR platform is an array of homogeneous, small footprint, RISC processors, each connected to its four nearest neighbors. No global control exists, yet it is distributed among the cores. Two versions are designed for the platform, along with a user-familiar programming model. A software version, CEDAR-S, is the basic implementation where adjacent cores are connected to each other via shared buffers. A co-processor called DMC (Direct Management of Communications) is added in the CEDAR-H version, to optimize the routing protocol. The DMCs are interconnected in a mesh fashion. Two novel concepts are proposed to enhance the adaptiveness of CEDAR. First, a distributed dynamic routing strategy, based on a bio-inspired algorithm, handles routing in a non-supervised fashion, and is independent of the physical placement of communicating tasks. The second concept presents dynamic distributed task migration in response to several system and application requirements. Results show that CEDAR scores high performances with its optimized routing strategy, compared to state-of-art networks. The migration cost is evaluated and adequate protocols are presented. CEDAR is shown to be a promising design concept for future manycores
Belabbas, Aïssame. "Conception d'un système de contrôle-commande reconfigurable pour l'assistance technique aux personnes handicapées." Lorient, 2007. http://www.theses.fr/2007LORIS090.
Повний текст джерелаZerkaoui, Salem. "Commande neuronale adaptative des systèmes non linéaires." Le Havre, 2007. http://www.theses.fr/2007LEHA0010.
Повний текст джерелаThe main contribution of this work is to propose a robust stable self-adaptive InDirect neural Network Control "IDNC" to control a broad variety of unknown linear, nonlinear, SISO and MIMO systems. The control scheme is made of an adaptive instantaneous neural model, a neural controller and an on-line parameter updating law. The IDNC parameters start at zero initial conditions which ensure that the performances do not depend on the initialization phase. Closed loop performances as well as sufficient conditions for asymptotic stability and robustness are derived from the Lyapunov approach. The simulations and experimental tests are carried out in order to validate the performances of the proposed approach. In particular, our contribution is used for the control of the Tennessee Eastman Challenge Process and a medical robot. Also, the proposed structure can easily be implemented in several practical applications
Enembreck, Fabricio. "Contribution à la conception d'agents assistants personnels adaptatifs." Compiègne, 2003. http://www.theses.fr/2003COMP1478.
Повний текст джерелаLn this thesis we study the developpement of personnal assistant agents. We have developped some mecanisms allowing the implementation of the conceptual models proposed by Ramos (2000). We stress the importance of the adaptativity of these models to the user's actions. During the developpement of the work we perceived that common methods of adaptation are not good enough for adaptative personal assistant agents. Thus, we introduced a new leaming approach (ELA) that has been evaluated and whose results are quite good. Our work ends up with a study allowing the integration of the different representations developed in the prototypes. We have designed a multi-agent representation for managing the different dynamic representations of the user with distributed leaming agents. This representation is modeled by a generic personalized multi-agent system architecture (SMAP) solving the problems introduced by Ramos' centralized architecture and allowing personalization
Garcia, Samuel. "Architecture reconfigurable dynamiquement a grain fin pour le support d'un système d'exploitation temps réel." Paris 6, 2012. http://www.theses.fr/2012PA066495.
Повний текст джерелаMost of anticipated future applications share four major characteristics. They might all require an increased computing capacity, they will implies to take real time into account, they represent a big step in terms of complexity compared with todays typical applications, and will have to deal with the dynamic nature of the real physical world. Fine grained dynamically reconfigurable architecture (FGDRA) can be seen as next evolution of today's FPGA, aiming at dealing with very dynamic and complex real time applications while providing comparable potential computing power due to the possibility to fine tune execution architecture at a fine grain level. To make this kind of devices usable for real application designer complexity has to be abstracted by an operating system layer and adequate tool set. This combination would form an adequate solution to support future applications. This thesis exposes an innovative FGDRA architecture called OLLAF. This architecture answer both technical issues on reconfigurable computing and practical problematics of application designers. The whole architecture is designed to work in symbiosis with an operating system. Studies presented here will more particularly focus on hardware task management mechanisms in a preemptive system. We will first present our work toward trying to implement such mechanisms using existing FPGA and show that those existing architectures have to evolve to efficiently support an operating system in a highly dynamic real time situation. The OLLAF architecture will then be explained and the hardware task management mechanism will be highlighted. We then present two studies that prove this approach to constitute a huge gain compared with existing platforms in terms of resulting operating system overhead even for static application cases where dynamical reconfiguration is used only for computing resource sharing. For highly dynamical real time cases we show that not only it could lower the overhead, but it will also support cases that existing devices just cannot support
Hoinville, Thierry. "Evolution de contrôleurs neuronaux plastiques : de la locomotion adaptée vers la locomotion adaptative." Versailles-St Quentin en Yvelines, 2007. http://www.theses.fr/2007VERS0022.
Повний текст джерелаLes recherches menées en robotique évolutionniste se sont avant tout focalisées sur l'évolution de contrôleurs neuronaux structurellement et paramétriquement figés, pour la locomotion de robots qui roulent, marchent, nagent ou volent. Cette démarche a permis la conception de contrôleurs bien adaptés à des environnements constants, mais non adaptatifs aux variations de ceux-ci. Pour y remédier, certains roboticiens ont suggéré de faire évoluer des neuro-contrôleurs non plus figés, mais plastiques. Notre approche s'inscrit dans ce revirement et vise à ce que les robots à pattes puissent adapter leur locomotion aussi bien aux perturbations extérieures, qu'aux éventuelles détériorations de leurs structures mécaniques. Nous proposons en premier lieu une revue des phénomènes de plasticiténeuronale et de leur modélisation, destinée essentiellement aux roboticiens. Nous dressons ensuite un état de l'art de l'évolution de neuro-contrôleurs plastiques et critiquons la plausibilité biologique des modèles développés. Notre première contribution s'inspire des travaux de la robotique évolutionniste et aborde le dilemme de l'évolution de contrôleurs à la fois flexibles et stables. Ainsi, nous employons des contraintes homéostatiques pour stabiliser la plasticité de contrôleurs assurant la locomotion d'un robot monopode confronté à une perturbation freinant son avancée. Notre deuxième contribution s'inspire des connaissances acquises sur les générateurs centraux de pattern (CPG) et leur plasticité. Ainsi,nous proposons l'évolution d'oscillateurs à relaxation soumis à neuromodulation pour la locomotion adaptative d'un robot myriapode confronté à d'éventuelles amputations de pattes
Charra, Olivier. "Conception de noyaux de systèmes embarqués reconfigurables." Grenoble 1, 2004. http://www.theses.fr/2004GRE10047.
Повний текст джерелаThe vision of the emergence of a global environment for the information management where most of the physical object around us will be equipped with processors, communication capabilities and interconnected through various networks forces us to redesign the computing systems. Instead of heavy, monolithic and non-evolutive systems, we must design light, flexible and reconfigurable systems. This work presents a new architecture allowing the conception and development of flexible and reconfigurable operating system kernels for embedded systems
Guillet, Sébastien. "Modélisation et contrôle de la reconfiguration : application aux systèmes embarqués dynamiquement reconfigurables." Lorient, 2012. http://www.theses.fr/2012LORIS275.
Повний текст джерелаBruguier, Florent. "Méthodes de caractérisation et de surveillance des variations technologiques et environnementales pour systèmes reconfigurables adaptatifs." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2012. http://tel.archives-ouvertes.fr/tel-00965377.
Повний текст джерелаMovahedi, Zeinab. "Une architecture autonomique pour les réseaux sans-fil : proposition et méthodologie d'évaluation." Paris 6, 2011. http://www.theses.fr/2011PA066537.
Повний текст джерелаGreff, Florian. "SDRN : réseau maillé temps réel dynamique défini par logiciel." Electronic Thesis or Diss., Université de Lorraine, 2018. http://www.theses.fr/2018LORR0076.
Повний текст джерелаWe are studying a new kind of embedded real-time mesh network. Mesh networking of the components of embedded systems reconciles their real-time constraints with the new application needs in terms of bandwidth and tight interactions. The plurality of communication paths results in increased flexibility, resilience, scalability and load balancing characteristics. However, this requires the ability to dynamically allocate network resource with respect to the needs of running applications. Our approach is to allow applications to make online real-time flow resource requests and consequently allot network resources according to these requirements. To this end, we have designed the Software-Defined Real-time Networking (SDRN) architecture. It addresses flow isolation, timing analysis, routing, fault tolerance, as well as the interfaces with the application layer and the lower layers of the system. It also allows any module to be replaced without interfering with the remainder of the architecture. It has been validated via an implementation on an in-silicon RapidIO platform. This thesis describes our research on the SDRN architecture. It also proposes an original method for experimenting on embedded networks, ERICA. The ERICA framework automatically generates all what is needed to conduct a network experiment in a selected environment (such as a simulator or a testbed), with both physical and simulated aspects. Hence, it allows the researcher to perform a high-level thinking of the whole experimentation process and to reuse applications and experiment designs from an experimentation stack to another
Debroucke, Romain. "Développements de capacités variables en technologie silicium pour les applications RF et mmW." Thesis, Lille 1, 2011. http://www.theses.fr/2011LIL10135.
Повний текст джерелаMultiplication of standards and communication protocols allows nowadays a wider range of services for each consumer. If this multiplication of standards is more significant for users, it creates an increased complexity of Rx/Tx communication modules both in RF (3G, 4G), or mmW (W-HDMI, imaging, radar, sensors ) range. These systems Rx/Tx must be able nowadays to support multiple frequency bands (tri-band mobilephone as an example), or agile functions generation, it is called reconfigurable systems. In this context, that this thesis work taking place which aims to address the challenge of high performances tunable capacitances development in RF and mmW range. The first chapter entitled "Emergence of the need for high performances non linear capacitances in silicon technology" identifies the limitations of usual tunable capacitances offer at STMicroelectronics. The second chapter " AttoFarad capacitance characterization: The Challenge of high impedance measurement" discuss the problem of characterization of very small capacity, called attoFarad, used mainly in digital controlled oscillator (DCO). Nowadays these aF capacitance are very difficult to modelize regarding problems of standard measurement methods. We will discuss of a new method for high frequency characteriziation with a relative precision of 50aF 1 to 10GHz. The third chapter entitled "From analog to digital, developments of tunable capacitances up to millimeter frequencies in CMOS and BiCMOS technologies," will focus on figures of merit improvements of the capacitance (tuning ration, resonant frequency, linear law...) through optimization of existing solution (MOS varactor) and the design of new digital architectures. The final chapter "Performances evaluation through referent mmW circuits " exhibit, the design, realization and characterization of two agile functions, a phase shifter and a in-situ tuner, both in the mmW range. To conclude, all these thesis work provide new solutions for the design of RF and mmW circuits. They have extended the tunable capacitances offer at STMicroelectronics in advanced technologie nodes but also to investigate and validate a new concept of characterization necessary for all of these developments
Dirani, Mariana. "Resource allocation and son based radio resource management in cellular and wireless networks." Paris 6, 2011. http://www.theses.fr/2011PA066480.
Повний текст джерелаBouguelia, Mohamed-Rafik. "Classification et apprentissage actif à partir d'un flux de données évolutif en présence d'étiquetage incertain." Electronic Thesis or Diss., Université de Lorraine, 2015. http://www.theses.fr/2015LORR0034.
Повний текст джерелаThis thesis focuses on machine learning for data classification. To reduce the labelling cost, active learning allows to query the class label of only some important instances from a human labeller.We propose a new uncertainty measure that characterizes the importance of data and improves the performance of active learning compared to the existing uncertainty measures. This measure determines the smallest instance weight to associate with new data, so that the classifier changes its prediction concerning this data. We then consider a setting where the data arrives continuously from an infinite length stream. We propose an adaptive uncertainty threshold that is suitable for active learning in the streaming setting and achieves a compromise between the number of classification errors and the number of required labels. The existing stream-based active learning methods are initialized with some labelled instances that cover all possible classes. However, in many applications, the evolving nature of the stream implies that new classes can appear at any time. We propose an effective method of active detection of novel classes in a multi-class data stream. This method incrementally maintains a feature space area which is covered by the known classes, and detects those instances that are self-similar and external to that area as novel classes. Finally, it is often difficult to get a completely reliable labelling because the human labeller is subject to labelling errors that reduce the performance of the learned classifier. This problem was solved by introducing a measure that reflects the degree of disagreement between the manually given class and the predicted class, and a new informativeness measure that expresses the necessity for a mislabelled instance to be re-labeled by an alternative labeller
Petit, Dorine. "Unification des stratégies de contrôle de réseau embarqué temps-réel reconfigurable." Electronic Thesis or Diss., Université de Lorraine, 2018. http://www.theses.fr/2018LORR0164.
Повний текст джерелаThis thesis is part of a long-lasting relationship between the CRAN and the CNES since 10 years. This involves studying the pertinence of replacing the buses currently embedded in European’s launchers, by off-the-shelf switched standard Ethernet architecture. Previous works qualified Ethernet regarding two quality of service items: real time and reliability. We continue this work by validating these first results on a platform, true scale architecture reproduction which will be embedded in launchers. Then, we work on an original quality of service item, typical of the spatial requirements, called observability. The objective is to re-play the whole flight, afterwards, on ground, knowing that the launcher will not come back, and the architecture is segmented, conferring to the network the dimension of non synchronized multi-domains of observation. The synthesis of all the captures implies to guarantee that the order and the freshness of launcher control information are consistent with the specified applicative operation, and that what is observed is compliant to what really happened. Finally, we propose an over-sampling which is dynamically regulated in order to benefit from the available bandwidth without disturbing real time, reconfiguration and observability requirements
Marques, Nicolas. "Méthodologie et architecture adaptative pour le placement efficace de tâches matérielles de tailles variables sur des partitions reconfigurables." Electronic Thesis or Diss., Université de Lorraine, 2012. http://www.theses.fr/2012LORR0139.
Повний текст джерелаFPGA-based reconfigurable architectures can deliver appropriate solutions for several applications as they allow for changing the performance of a part of the FPGA while the rest of the circuit continues to run normally. These architectures, despite their improvements, still suffer from their lack of adaptability when confronted with applications consisting of variable size material tasks. This heterogeneity may cause wrong placements leading to a sub-optimal use of resources and therefore a decrease in the system performances. The contribution of this thesis focuses on the problematic of variable size material task placement and reconfigurable region effective generation. A methodology and an intermediate layer between the FPGA and the application are proposed to allow for the effective placement of variable size material tasks on reconfigurable partitions of a predefined size. To approve the method, we suggest an architecture based on the use of partial reconfiguration in order to adapt the transcoding of one video compression format to another in a flexible and effective way. A study on the reconfigurable region partitioning for the entropy encoder material tasks (CAVLC / VLC) is proposed in order to show the contribution of partitioning. Then an assessment of the gain obtained and of the method additional costs is submitted
Leduc, Guilain. "Performance et sécurité d'une Blockchain auto-adaptative et innovante." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0220.
Повний текст джерелаResearch on blockchain application frameworks rarely offers performance evaluation. This thesis proposes a comprehensive methodology to help software integrators better understand and measure the influence of configuration parameters on the overall quality of long-term service performance. In order to improve performance, the new adaptive consensus protocol Sabine (Self-Adaptive BlockchaIn coNsEnsus) is proposed to dynamically modify one of these parameters in the PBFT consensus. The configuration parameter of this consensus is the number of validators involved and result of a trade-off between security and performance. The Sabine protocol maximises this number provided that the output rate matches the input rate. Sabine is evaluated and validated in real-world settings, the results of which show that Sabine has an acceptable relative error between the requested and committed transaction rates. Two new validator selection algorithms are proposed that reverse the random paradigm of current protocols to select the nodes leading to better performance. The first is based on a reputation system that rewards the fastest nodes. The second selects the closest nodes by imposing a continuous rotation of the selection. These two algorithms have been simulated and their impact on decentralisation discussed. This selection, associated with Sabine, improves security by giving the system more margin to increase the number of validators. This work opens the way to more reactive chains, with less latency and more throughput
Hamdoun, Safa. "Algorithmes adaptatifs et efficaces de partage de ressources radio pour les communications de type MTC : cas de coexistence H2H/M2M." Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1096/document.
Повний текст джерелаThe Internet-of-Things (IoT) refers to the ever-growing network of everyday objects that interconnect to each other or to other Internet-enabled systems via wireless sensors attached to them. IoT envisions a future where billions of smart devices will be connected and managed through a range of communication networks and cloud-based servers, enabling a variety of monitoring and control applications. Machine-to-Machine (M2M) communications supported by cellular networks, also known as Machine-Type-Communications (MTC) acts as a key technology for partially enabling IoT. M2M communications is a new technology paradigm that facilitates the ubiquitous connectivity between a myriad of devices without requiring human intervention. The surge in the demand for connectivity has further challenged network operators to design novel radio resource allocation algorithms at affordable costs to handle the massive scale of MTC.Different from current radio access technologies tailored to traditional Human- to-Human (H2H) communications, the goal of this thesis is to provide novel efficient and adaptive radio resource sharing schemes for MTC under a H2H/M2M coexistence scenario. We first provide a suitable multiple access scheme to address the joint spectrum scarcity, scalability and Base Station (BS) overload issues. Toward this end, we design a group-based operation where MTC corresponds to local uplink communications between Machine-Type-Devices (MTDs), which represent a specific type of devices that do not rely on the presence of a human interface, and a Machine-Type-Head (MTH). This latter plays the role of a cluster head that relays the information to the BS. We thus address the need to aggregate M2M and Device-to-Device (D2D) technology, as one of the major components of the future evolving cellular networks. Having said that, we first propose in this thesis to model the radio resource sharing problem between MTDs and H2H users as a bipartite graph and develop a novel interference-aware graph-based radio resource sharing algorithm for MTC so as to mitigate the co-channel interference and thus enhance network efficiency. Moreover, low-complexity semi-distributed solution is investigated to alleviate the communication overhead of a centralized solution that we propose as well. Then, as a second contribution, we examine how M2M devices can share the available radio resources in cellular networks with no or limited impact on existing H2H services. Consequently, we propose a joint spectrally and power efficient radio resource sharing scheme. Convinced by the strength of the bipartite graph modeling for the resource sharing problem between H2H users and M2M devices, we empower the graph-based radio resource sharing algorithm with a novel adaptive power control feature using one of two following mechanisms: the Proportional Integral Derivative (PID) controller and the fuzzy logic. Finally, in our third contribution of this thesis, we develop a power efficient and fully-distributed radio resource sharing framework for MTC underlying cellular networks. We use game theory and model the resource sharing problem as an efficient hybrid-game where M2M devices compete for radio resources and switch opportunistically, as M2M devices are selfish in nature, between non-cooperative and cooperative games. The different derived solutions are extended to existing cellular networks, and extensive simulation studies in the context of LTE are conducted. The various simulation results show that the proposed solutions can significantly increase the efficiency of the spectrum usage, mitigate the negative effect on H2H services and save the battery life of M2M devices
Peralta, Cortes Héctor. "Création de critères d'évaluation de dispositifs d'assistance pour personnes en situation de handicap." Versailles-St Quentin en Yvelines, 2009. http://www.theses.fr/2009VERS0026.
Повний текст джерелаPLEIA project is a methodological platform of different software tools for the evaluation of different interfaces for persons with disabilities. This platform is based on different modules: PLEIA Functional awaking which is used to study the interactions between the interfaces peripheral and awaking systems (toys and robots), PLEIA Exercise which is based on the pointing devices for a personal computer, PLEIA Functional which is oriented to the command interfaces for the technical aids such as electric powered wheelchair. These modules are used by occupational therapists to create a basic evaluation. It tries to create tests with the integration of interfaces and performance analysis of patients. This thesis is based on PLEIA Exercise and Evaluation modules. The experimental results are shown for myopathic Children. The obtained results are promising
Allègre, Olivier. "Adapting the Prerequisite Structure to the Learner in Student Modeling." Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS116.
Повний текст джерелаData-driven learner models aim to represent and understand students' knowledge and other meta-cognitive characteristics to support their learning by making predictions about their future performance. Learner modeling can be approached using various complex system models, each providing a different perspective on the student and the learning process. Knowledge-enhanced machine learning techniques, such as Bayesian networks, are particularly well suited for incorporating domain knowledge into the learner model, making them a valuable tool in student modeling.This work explores the modeling and the potential applications of a new framework, called E-PRISM, for Embedding Prerequisite Relationships In Student Modeling, which includes a learner model based on dynamic Bayesian networks. It uses a new architecture for Bayesian networks that rely on the clause of Independence of Causal Influences (ICI), which reduces the number of parameters in the network and allows enhanced interpretability. The study examines the strengths of E-PRISM, including its ability to consider the prerequisite structure between knowledge components, its limited number of parameters, and its enhanced interpretability. The study also introduces a novel approach for approximate inference in large ICI-based Bayesian networks, as well as a performant parameter learning algorithm in ICI-based Bayesian networks. Overall, the study demonstrates the potential of E-PRISM as a promising tool for discovering the prerequisite structure of domain knowledge that may be adapted to the learner with the perspective of improving the outer-loop adaptivity
Robert, Fabien. "Solution de filtrage reconfigurable en technologie CMOS 65nm pour les architectures d'émission numériques." Thesis, Paris Est, 2011. http://www.theses.fr/2011PEST1046/document.
Повний текст джерелаThis thesis addresses the technical and technological challenges in the design of “all digital” reconfigurable mobile architectures operating cellular standard bands (GSM, WCDMA, HSUPA and LTE). With the ever-changing communication needs, mobile devices must be able to address different standards from a common architecture depending on free frequency bands, data rate and spectral constraints. In order to reduce costs, consumption and to obtain a greater integration, new architectures were developed and called multi-standard allowing a single transmitter to transmit each standard instead of parallelizing several radio architectures each dedicated to a particular standard. For several years nanoscale technologies such as 90nm or 65nm CMOS have emerged, clearing the way to replace analog functional blocks by greater digital functional blocks. In this study, we identify possible changes between "analog world" and "digital world" to move the digital boundary from the baseband to power amplifier. Several architectures have been studied with progressive digitization degrees to meet "all digital" architecture, comprising part of the power amplifier. Extensive work on the study of different cellular standards conducted jointly with the implementation and simulation of these architectures, let us identified the different technological and functional locks in the development of "all digital" architectures. Oversampling spurious constraints have emerged as dimensioning. For each band of each standard, these constraints were evaluated to define an optimization method of over-sampling frequency. However an external filter is required. A second step led us to identify and design a reconfigurable bandpass filtering technique for cellular bands from 1710 to 1980MHz with at least 60MHz of bandwidth in order to address the LTE, and 23dB attenuation at 390MHz from the center of the filter to address the most constringent filtering cases (bands 1, 3 and 10 in W-CDMA). We then designed and implemented a reconfigurable filter based on active inductors to ensure reconfigurability and very low insertion loss. This thesis permit from an actual architecture system issue and through a process to identify limitations of “all digital” architectures, to propose an adapted filtering solution. This filter was designed in 65nm CMOS, implemented. Measured performance is consistent with requirements
Kermorgant, Olivier. "Fusion d'informations multi-capteurs en asservissement visuel." Phd thesis, Université Rennes 1, 2011. http://tel.archives-ouvertes.fr/tel-00747052.
Повний текст джерелаGuzzo, Natale. "Facing the real challenges in wireless sensor network-based applications : an adaptative cross-layer self-organization WSN protocol." Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10190.
Повний текст джерелаWireless Sensor Networks (WSN) is one of the protagonists contributing to the evolution and the development of the Internet of Things (IoT). Several use cases can be found today in the different fields of the modern technology including the container shipping industry where containerized cargo accounts for about 60 percent of all world seaborne trade. In this context, TRAXENS developed a battery-powered device named TRAX-BOX designed to be attached to the freight containers in order to track and monitor the shipping goods along the whole supply chain. In this thesis, we present a new energy-efficient self-organizing WSN protocol stack named TRAX-NET designed to allow the TRAX-BOX devices to cooperate to deliver the sensed data to the TRAXENS platform.The results of simulations and field tests show that TRAX-NET well perform in the different scenarios in which it is supposed to operate and better fulfil the requirements of the assumed application in comparison with the existing schemes
Burger, Brice. "Fusion de données audio-visuelles pour l'interaction homme-robot." Phd thesis, Toulouse 3, 2010. http://thesesups.ups-tlse.fr/916/.
Повний текст джерелаIn the framework of assistance robotics, this PHD aims at merging two channels of information (visual and aiditive potentially avaible on a robot. The goal is ton complete and/ or conf rm data that an only channel could have supplied in order to perform advanced intatraction which goal is to interpret jointly speech gesture, in particular for the use of spatial references. In this thesis, we first de cribe the speech part of this work which consists in an embedded recognition and interpretation system for continuous speech. Then comes the vision part which is composed of a visual multi-target tracker that tracks, in 3D the head and the two hands of a human in front of the robot, and a second tracker for the head orientation. The outputs of these trackers are used to feed the gesture recognitive system described later. We continue with the description of a module dedicated to the fusion of the data outputs of these information sources in a probailistic framework. Last, we demonstrate the interest and feasibility of such a multimodal interface through some demonstrations on the LAAS-CNRS robots. All the modules described in this thesis are working in quasi-real time on these real robotic platforms
Farhat, Salman. "Safe Dynamic Reconfiguration of Applications with Features." Electronic Thesis or Diss., Université de Lille (2022-....), 2024. http://www.theses.fr/2024ULILB014.
Повний текст джерелаCloud applications and cyber-physical systems require frequent reconfiguration at run-time to adapt to changing needs and requirements, highlighting the importance of dynamic reconfiguration capabilities. Additionally, the environment platforms can extend and modify their services at run-time, which necessitates a compositional approach to allow the modifications of the configurations. To manage the variability of large systems' architecture, feature models are widely used at design-time with several operators defined to allow their composition. Existing approaches compute new valid configurations either at design time, at runtime, or both, leading to significant computational or validation overheads for each reconfiguration step. In addition, building correct-by-construction formal models to handle application reconfigurations is a complex and error-prone task, and there is a need to make it automated as far as possible.To address these challenges, we propose an approach named FeCo4Reco that leverages feature models to automatically generate, in a component-based formalism called JavaBIP, component-based run-time variability models that respect the feature model constraints. These component-based run-time variability models are executable and can be used at runtime to enforce the variability constraints, that is, to ensure the (partial) validity of all reachable configurations.As complex systems' architectures may evolve at run-time by acquiring new functionalities while respecting new constraints, we define composition operators for component-based run-time variability models that not only encode these feature model composition operators, but also ensure safe run-time reconfiguration. To prove the correctness and compositionality properties, we propose a novel multi-step UP-bisimulation equivalence and use it to show that the component-based run-time variability models preserve the semantics of the composed feature models.For the experimental evaluation, we demonstrated the applicability of our approach in real-world scenarios by generating a run-time model based on the feature model of the Heroku cloud platform using our approach. This model is then used to deploy a real-world web application on the Heroku platform. Furthermore, we measured the time and memory overheads induced by the generated run-time models on systems involving up to 300 features. The results show that the overheads are negligible, demonstrating the practical interest of our approach
Cudennec, Loïc. "CoRDAGe : un service générique de co-déploiement et redéploiement d’applications sur grilles." Phd thesis, Rennes 1, 2009. ftp://ftp.irisa.fr/techreports/theses/2009/cudennec.pdf.
Повний текст джерелаComputer grids are made of thousands of heterogeneous physical resources that belong to different administration domains. This makes the use of the grid very complex, especially for deploying distributed applications at large scale, while taking care of the dynamicity of the infrastructure. The contribution of the PhD Thesis is to propose a model that allows to dynamically and transparently deploy distributed applications. This model can be adapted, in a non-intrusive way, to handle different kind of applications. In order to validate our model, we propose an architecture named CoRDAGe. This leads to a prototype that can handle several applications, such as the JuxMem data-sharing service and the Gfarm distributed file system. Experimentations have been conducted within the Grid'5000 project
Vie, Jill-Jênn. "Modèles de tests adaptatifs pour le diagnostic de connaissances dans un cadre d'apprentissage à grande échelle." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLC090/document.
Повний текст джерелаThis thesis studies adaptive tests within learning environments. It falls within educational data mining and learning analytics, where student educational data is processed so as to optimize their learning.Computerized assessments allow us to store and analyze student data easily, in order to provide better tests for future learners. In this thesis, we focus on computerized adaptive testing. Such adaptive tests which can ask a question to the learner, analyze their answer on the fly, and choose the next question to ask accordingly. This process reduces the number of questions to ask to a learner while keeping an accurate measurement of their level. Adaptive tests are today massively used in practice, for example in the GMAT and GRE standardized tests, that are administered to hundreds of thousands of students. Traditionally, models used for adaptive assessment have been mostly summative : they measure or rank effectively examinees, but do not provide any other feedback. Recent advances have focused on formative assessments, that provide more useful feedback for both the learner and the teacher ; hence, they are more useful for improving student learning.In this thesis, we have reviewed adaptive testing models from various research communities. We have compared them qualitatively and quantitatively. Thus, we have proposed an experimental protocol that we have implemented in order to compare the most popular adaptive testing models, on real data. This led us to provide a hybrid model for adaptive cognitive diagnosis, better than existing models for formative assessment on all tried datasets. Finally, we have developed a strategy for asking several questions at the beginning of a test in order to measure the learner more accurately. This system can be applied to the automatic generation of worksheets, for example on a massive online open course (MOOC)
Ameer, Muhammad. "Integrated and multi-criteria approaches for process plan generation in reconfigurable manufacturing systems with consideration of system capabilities and product constraints." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0242.
Повний текст джерелаModern manufacturing systems are going through a paradigm shift where the focus is on the integrating the digital technologies in the production systems to address the challenge of uncertain market demands. Manufacturing systems needs certain amount responsiveness to address these uncertainties by adapting accordingly, and require more changeability at physical as well as logical levels. For this purpose, modern-day manufacturing systems are designed with dynamic resource capabilities, with modular components, so that they can provide the required amount of reconfigurability. From the perspective of "industry 4.0", reconfigurability is vital for the effective adaptation of manufacturing systems in a complex environment. Reconfigurability provides the quick adaptation of these systems along with quick responsiveness towards socio-techno-economic competitiveness. The objective is to respond to modern-day challenges (both external and internal), i.e. mass customization, globalization, product variety management, system reconfiguration management, and reducing the lead time.In this thesis, the design problem of reconfigurable manufacturing systems (RMS) is considered which meets the aforementioned requirements. The goal is to design a responsive system based on two key features modularity and reconfigurability. We study the RMS design problem as, the development of a process plan for a particular part of the part family along with the selection of the system's dynamic resource capabilities to perform that part. This work is divided into three parts: (1) Co-generation of process and setup plan for a part in the reconfigurable environment. The main objective is to develop a new approach to jointly consider the setup and process plan constraints, with consideration of relationships between the operations. (2) Minimisation of reconfiguration effort in process design. We propose a novel performance index of the effort generated by the machines and fixtures reconfiguration, and part transfer. The objective is to ensure better responsiveness and high performance of the designed process plan. (3) Maximisation the utilization of dynamic resource capabilities of RMS design. We consider a RMS design problem for machine selection, where selection of different types of reconfigurable machine tools (RMTs) are carried out for performing the process plan of considered part
Debackere, Florian. "Conception d'interactions motivationnelles, adaptatives et mobiles pour le changement de comportement : application à la lombalgie." Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASG060.
Повний текст джерелаToday's technologies are known as persuasive or motivational, referring to technologies, applications or services designed to induce changes in the attitudes and behaviors of those who use them. This is the subject of research in Human-Computer Interaction, in conjunction with theories from psychology linked, for example, to behavior change or motivation. Research on these technologies suggests that, in order to encourage long-term adherence, these “virtual coaches” need to be personalized and/or adapted according to users' individual characteristics (stage of behavior change, motivations, preferences, barriers). For example, according to Self-Determination Theory (SDT), an individual may present different forms of motivation, more or less effective. The aim is to identify the forms of motivation present in users, in order to offer services and interactions that reinforce or develop them. The use of such systems in the healthcare field has the potential to induce and reinforce health behaviors that are sometimes difficult for caregivers to establish. Offering personalized daily care represents a considerable human and financial cost for healthcare professionals. A mobile application has the advantage of being able to cope with these constraints. The aim of this thesis is to design this computer coach with and for patients suffering from chronic low back pain, to help them manage their condition, particularly with regard to their pain and the practice of regular physical activity. The major challenges of this thesis include the selection and coordination of different existing methodologies, as well as the evaluation of the application's quality and effectiveness. Key issues address a thorough understanding of patients in terms of psychological characteristics, the implementation of a user-centered and participatory approach to meet patients' needs, the tailoring of the intervention to foster the behavior change process, and the evaluation of this intervention on patients' engagement and behavior change process. The contributions are structured around three axes: (1) the proposal of a design and evaluation metamodel for adaptive digital interventions to support behavior change, (2) the identification of low back pain patient profiles based on psychological characteristics related to the behavior change process, and (3) the design of a mobile application for low back pain patients. Thus, this thesis proposes a useful approach to guide the design of adaptive human-machine interventions and interactions, based on theories in the psychology of behavior change and involving professionals and patients in the design process
Kaushik, Rituraj. "Data-Efficient Robot Learning using Priors from Simulators." Electronic Thesis or Diss., Université de Lorraine, 2020. http://www.theses.fr/2020LORR0105.
Повний текст джерелаAs soon as the robots step out in the real and uncertain world, they have to adapt to various unanticipated situations by acquiring new skills as quickly as possible. Unfortunately, on robots, current state-of-the-art reinforcement learning (e.g., deep-reinforcement learning) algorithms require large interaction time to train a new skill. In this thesis, we have explored methods to allow a robot to acquire new skills through trial-and-error within a few minutes of physical interaction. Our primary focus is to incorporate prior knowledge from a simulator with real-world experiences of a robot to achieve rapid learning and adaptation. In our first contribution, we propose a novel model-based policy search algorithm called Multi-DEX that (1) is capable of finding policies in sparse reward scenarios (2) does not impose any constraints on the type of policy or the type of reward function and (3) is as data-efficient as state-of-the-art model-based policy search algorithm in non-sparse reward scenarios. In our second contribution, we propose a repertoire-based online learning algorithm called APROL which allows a robot to adapt to physical damages (e.g., a damaged leg) or environmental perturbations (e.g., terrain conditions) quickly and solve the given task. In this work, we use several repertoires of policies generated in simulation for a subset of possible situations that the robot might face in real-world. During the online learning, the robot automatically figures out the most suitable repertoire to adapt and control the robot. We show that APROL outperforms several baselines including the current state-of-the-art repertoire-based learning algorithm RTE by solving the tasks in much less interaction times than the baselines. In our third contribution, we introduce a gradient-based meta-learning algorithm called FAMLE. FAMLE meta-trains the dynamical model of the robot from simulated data so that the model can be adapted to various unseen situations quickly with the real-world observations. By using FAMLE with a model-predictive control framework, we show that our approach outperforms several model-based and model-free learning algorithms, and solves the given tasks in less interaction time than the baselines
Foulonneau, Anthony. "Les technologies persuasives adaptatives." Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAM085/document.
Повний текст джерелаThis thesis deals with persuasive technologies, and in particular adaptation of the persuasion in order to optimize efficiency and relevance of those technologies. Persuasive technologies are technologies design to change behaviors without using coercion or deception. Numerous techniques to shape user behavior but few and perfectible methods to design these technologies characterized the research domain of persuasive technologies.The background of persuasive technologies is the traditional interpersonal persuasion, studied for over two thousand years in rethoric, philosophy, and more recently psychology. This last discipline offers many theories and models to understand more precisely the process that determine human behaviors. These theories show in particular that persuasive situations are complex, varied, with many influence factors. That is why we propose the notion of adaptive persuasive technologies : technologies able to adapt their persuasive stategies to the user context. To design these products and services, we propose in the first place a model of the persuasive context, that is all the constraints that influence the practice of a targeted behavior by the user at a given time. Each constraint in the persuasive context is at the same time an adaptation criteria and an action lever for the adaptive persuasive technology. For each lever, we identify the persuasive principle that can be used to move it. In a second time, we propose a problem space that characterized the adaptation of the persuasion, thanks to five axis : the purpose, the target, the criteria and the dynamic of the adaptation, and the user role in the adaptation process. Finally, by making and assessing TILT, a persuasive application dedicated to smartphone usage regulation, we show that the adaptation of the persuasion, with the use of the persuasive contexte model and the problem space, benefits to the persuasive efficiency
Kebbati, Mohammed-Yassine. "Learning-based reinforced control strategies for autonomous vehicles." Electronic Thesis or Diss., université Paris-Saclay, 2023. https://www.biblio.univ-evry.fr/theses/2023/interne/2023UPAST071.pdf.
Повний текст джерелаIn a world where everyone is constantly on the move and has more preferences for personal mobility, the number of cars on the road is constantly increasing. This has induced higher risks for road accidents, degraded traffic conditions, and further aggravated air pollution. Consequently, the research community has been working toward a shift to autonomous driving for decades. The latter can reshape mobility and transportation by reducing road accidents, traffic jams, and air pollution, yielding energy efficiency, convenience, and more productivity as significant driving time will be used for other activities instead. Autonomous vehicles are complex systems consisting of several modules that perform perception, decision-making, planning, and control. The control module, consisting of longitudinal control for speed tracking and lateral control for path tracking, is essential for achieving automatic driving. Due to the highly dynamic and constantly changing nature of road environments, the control module in autonomous driving systems needs to learn and adapt to these dynamic environments by exploiting available data and using different learning techniques. This thesis contributes to the state-of-the-art enhanced control strategies applied to autonomous driving. The contributions address the longitudinal and the lateral control tasks separately and then the combined coordinated and coupled lateral and longitudinal control. For the longitudinal control, we propose the adaptive PID approach using two techniques : offline optimization and adaptation using genetic algorithms (GA-PID) and online learning and adaptation with neural networks (NNPID). We introduce an adaptive MPC technique enhanced with a new, improved PSO algorithm for steering control. Then we achieve MPC online parameter adaption by using neural networks (NNMPC) and adaptive neuro-fuzzy inference systems (ANFIS-MPC), which learn to adapt the controller to the changing working conditions and external disturbances. For the combined coordinated lateral and longitudinal control, we propose a PSO-PID to handle the task of speed tracking and an enhanced linear parameter varying model predictive controller (LPV-MPC) to control lateral dynamics. The LPV-MPC is enhanced with an improved cost function to provide better performance and stability and formulated with an adaptive LPV model, in which a recursive estimator estimates the tire cornering stiffness coefficients. Then, we address the coupled speed and steering control by developing a more elaborate LPV-MPC capable of simultaneously handling both lateral and longitudinal vehicle dynamics. Furthermore, a neural network adapts the controller's prediction model online, and an improved genetic algorithm further optimizes its cost function. Finally, we tackle a more challenging autonomous driving problem where the vehicle drives at its handling limits, namely autonomous racing. We introduce a real-time nonlinear model predictive controller (NMPC) coupled with a moving horizon estimator (MHE). We solve the optimal racing problem with an NMPC-based offline trajectory planner that computes the best trajectory while considering the physical limits of the vehicle and circuit constraints. Then, we further enhance the control strategy by adding a learning extension based on Gaussian process regression. The latter improves the NMPC predictions by learning and correcting the mismatch between the actual vehicle dynamics and the NMPC prediction model
Toldov, Viktor. "Adaptive MAC layer for interference limited WSN." Thesis, Lille 1, 2017. http://www.theses.fr/2017LIL10002/document.
Повний текст джерелаIn the era of the Internet of Things, the number of connected devices is growing dramatically. Often, connected objects use Industrial, Scientific and Medical radio bands for communication. These kinds of bands are available without license, which facilitates development and implementation of new connected objects. However, it also leads to an increased level of interference in these bands. Interference not only negatively affects the Quality of Service, but also causes energy losses, which is especially unfavorable for the energy constrained Wireless Sensor Networks (WSN). In the present thesis the impact of the interference on the energy consumption of the WSN nodes is studied experimentally. The experimental results were used to estimate the lifetime of WSN nodes under conditions of different levels of interference. Then, a Thompson sampling based Cognitive Radio (CR) adaptive solution is proposed and evaluated via both, simulation and hardware implementation. Results show that this approach finds the best channel quicker than other state of the art solutions. An extension for multihop WSN was proposed for this CR solution and evaluated by hardware implementation in the framework of EWSN Dependability Competition. Finally, an adaptive WildMAC MAC layer protocol is proposed for the usecase of the LIRIMA PREDNET wildlife animal tracking project. Obtained field range test data were used to theoretically estimate cell densities and deployment zone coverage in this Low Power Widea Area Network (LPWAN). Then performance of the protocol was evaluated in WSNet simulation. The results show performance that allows to respect PREDNET project requirements with the given coverage
Fragne, Didier. "Proposition de l'architecture de l'agent gestionnaire du modèle de l'apprenant dans un système tuteur multi-agents en apprentissage de la lecture : contribution au projet AMICAL." Phd thesis, Université Blaise Pascal - Clermont-Ferrand II, 2009. http://tel.archives-ouvertes.fr/tel-00449160.
Повний текст джерелаAouag, Sofiane. "Individualisation de l'apprentissage dans un Système Tuteur Intelligent : cas de l'apprentissage de la lecture dans un système AMICAL." Phd thesis, Université Blaise Pascal - Clermont-Ferrand II, 2010. http://tel.archives-ouvertes.fr/tel-00658846.
Повний текст джерелаBen, Jannet Mohamed Amer. "Évaluation adaptative des systèmes de transcription en contexte applicatif." Thesis, Université Paris-Saclay (ComUE), 2015. http://www.theses.fr/2015SACLS041/document.
Повний текст джерелаIt is important to regularly assess the technological innovation products in order to estimate the level of maturity reached by the technology and study the applications frameworks in which they can be used. Natural language processing (NLP) aims at developing modules and applications that automatically process the human language. That makes the field relevant to beth research and technological innovation. For years, the different technological modules from the NLP were developed separately. Therefore, the existing evaluation methods are in most modular. They allow to evaluate only one module at a time, while today, many applications need to combine several NLP modules to solve complex tasks. The new challenge in terms of evaluation is then to evaluate the different modules while taking into account the applicative context.Our work addresses the evaluation of Automatic Speech Recognition (ASR) systems according to the applicative context. We will focus on the case of Named Entities Recognition (NER) from spoken documents transcriped automatically. In the first part, we address the issue of evaluating ASR systems according to the application context through a study of the state of the art. We describes the tasks of ASR and NER proposed during several evalution campaigns and we discuss the protocols established for their evaluation. We also point the limitations of modular evaluation approaches and we expose the alternatives measures proposed in the literature. In the second part we describe the studied task of named entities detection, classification and decomposition and we propose a new metric ETER (Entity Tree Error Rate) which allows to take into account the specificity of the task and the applicative context during the evaluation. ETER also eliminates the biases observed with the existing metrics. In the third part, we define a new measure ATENE (Automatic Transcriptions Evaluation for Named Entities) that evaluates the quality of ASR systems and the impact of their errors for REN systems applied downstream. Rather than directly comparing reference and hypothesis transcriptions, ATENE measure how harder it becames to identify entities given the differences between hypothesis and reference by comparing an estimated likelihood of presence of entities. It is composed of two elementary measurements. The first aims to assess the risk of entities deletions and substitutions and the second aims to assess the risk of entities insertions caused by ASR errors.Our validation experiments show that the measurements given by ATENE correlate better than other measures from the state of the art with the performance of REN systems
Majed, Aliah. "Sensing-based self-reconfigurable strategies for autonomous modular robotic systems." Electronic Thesis or Diss., Brest, École nationale supérieure de techniques avancées Bretagne, 2022. http://www.theses.fr/2022ENTA0013.
Повний текст джерелаModular robotic systems (MRSs) have become a highly active research today. It has the ability to change the perspective of robotic systems from machines designed to do certain tasks to multipurpose tools capable of accomplishing almost any task. They are used in a wide range of applications, including reconnaissance, rescue missions, space exploration, military task, etc. Constantly, MRS is built of “modules” from a few to several hundreds or even thousands. Each module involves actuators, sensors, computational, and communicational capabilities. Usually, these systems are homogeneous where all the modules are identical; however, there could be heterogeneous systems that contain different modules to maximize versatility. One of the advantages of these systems is their ability to operate in harsh environments in which contemporary human-in-the-loop working schemes are risky, inefficient and sometimes infeasible. In this thesis, we are interested in self-reconfigurable modular robotics. In such systems, it uses a set of detectors in order to continuously sense its surroundings, locate its own position, and then transform to a specific shape to perform the required tasks. Consequently, MRS faces three major challenges. First, it offers a great amount of collected data that overloads the memory storage of the robot. Second it generates redundant data which complicates the decision making about the next morphology in the controller. Third, the self reconfiguration process necessitates massive communication between the modules to reach the target morphology and takes a significant processing time to self-reconfigure the robotic. Therefore, researchers’ strategies are often targeted to minimize the amount of data collected by the modules without considerable loss in fidelity. The goal of this reduction is first to save the storage space in the MRS, and then to facilitate analyzing data and making decision about what morphology to use next in order to adapt to new circumstances and perform new tasks. In this thesis, we propose an efficient mechanism for data processing and self-reconfigurable decision-making dedicated to modular robotic systems. More specifically, we focus on data storage reduction, self-reconfiguration decision-making, and efficient communication management between modules in MRSs with the main goal of ensuring fast self-reconfiguration process
Saïdi, Taofik. "Architectures matérielles pour la technologie WCDMA étendue aux systèmes mulit-antennes." Phd thesis, Université Rennes 1, 2008. http://tel.archives-ouvertes.fr/tel-00446060.
Повний текст джерелаToussaint, Marion. "Une contribution à l'industrie 4.0 : un cadre pour sécuriser l'échange de données standardisées." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0121.
Повний текст джерелаThe recent digital transformation of the manufacturing world has resulted in numerous benefits, from higher quality products to enhanced productivity and shorter time to market. In this digital world, data has become a critical element in many critical decisions and processes within and across organizations. Data exchange is now a key process for the organizations' communication, collaboration, and efficiency. Industry 4.0 adoption of modern communication technologies has made this data available and shareable at a quicker rate than we can consume or track it. This speed brings significant challenges such as data interoperability and data traceability, two interdependent challenges that manufacturers face and must understand to adopt the best position to address them. On one hand, data interoperability challenges delay faster innovation and collaboration. The growing volume of data exchange is associated with an increased number of heterogeneous systems that need to communicate with and understand each other. Information standards are a proven solution, yet their long and complex development process impedes them from keeping up with the fast-paced environment they need to support and provide interoperability for, slowing down their adoption. This thesis proposes a transition from predictive to adaptive project management with the use of Agile methods to shorten the development iterations and increase the delivery velocity, increasing standards adoption. While adaptive environments have shown to be a viable solution to align standards with the fast pace of industry innovation, most project requirements management solutions have not evolved to accommodate this change. This thesis also introduces a model to support better requirement elicitation during standards development with increased traceability and visibility. On the other hand, data-driven decisions are exposed to the speed at which tampered data can propagate through organizations and corrupt these decisions. With the mean time to identify (MTTI) and mean time to contain (MTTC) such a threat already close to 300 days, the constant growth of data produced and exchanged will only push the MTTI and MTTC upwards. While digital signatures have already proven their use in identifying such corruption, there is still a need for formal data traceability framework to track data exchange across large and complex networks of organizations to identify and contain the propagation of corrupted data. This thesis analyses existing cybersecurity frameworks, their limitations, and introduces a new standard-based framework, in the form of an extended NIST CSF profile, to prepare against, mitigate, manage, and track data manipulation attacks. This framework is also accompanied with implementation guidance to facilitate its adoption and implementation by organizations of all sizes