Thèse doctorat : Mathématiques : Strasbourg 1 : 2006. / Titre provenant de l'écran-titre. Bibliogr. 2 p.

Thesis (doctoral)--Universität Kaiserslautern, 2003. / Includes bibliographical references (p. [207]-212) and index.

No description available.

Le cœur est un organe central au fonctionnement du système cardiovasculaire. Il est physiologiquement compartimenté et est constitué de cellules spécialisées qui régulent les impulsions électriques ainsi que la contraction du myocarde. Le cœur adapte le flux sanguin en fonction des besoins du corps. En condition pathologique, le cœur recourt toutefois à des mécanismes compensatoires. Au niveau physiologique, la compensation s’observe par l’hypertrophie des cardiomyocytes qui, bien que bénéfique à court terme, exacerbe à long terme la fonction cardiaque. L’activation des « mitogen activated protein kinases » (MAPK) contribue autant au maintien de la fonction physiologique qu’à la détérioration pathologique du myocarde et serait également une cause de l’hypertrophie observée. Parmi les 5 groupes de MAPK connues, la MAPK p38 est formée de 4 isoformes dont les sérine/thréonine kinases p38α et p38γ sont exprimées de façon prédominante dans le cœur. Les p38 partagent les mêmes activateurs, mais leurs effecteurs diffèrent. Bien que le rôle de p38α semble impliqué dans l’aggravement des troubles cardiaques, celui de p38γ ne semble pas redondant à p38α et demeure incompris. Cette isoforme possède un motif de liaison aux domaines PDZ, unique chez les MAP kinases. Également, chez les cellules cardiaques, elle transloque au noyau en condition de stress. Le but de l’étude ici est de comprendre le rôle de p38γ et de ses motifs uniques sur la structure et la taille des cardiomyocytes. Afin de répondre au but de l’étude, plusieurs mutants adénoviraux de p38 ont été conçus. Un des mutants ne possède pas le motif de liaison aux domaines PDZ, deux autres contrôlent la localisation cellulaire soit au noyau, soit au cytoplasme, et un autre mutant est muté au site de phosphorylation. Des cardiomyocytes en culture ont été infectés par les différents mutants en présence de leur activateur en amont ou de la β-galactosidase. Les réseaux d’α-actinine, ainsi que la taille des cardiomyocytes, ont été observés par microscopie. Les observations effectuées montrent que p38γ entraîne une désorganisation des réseaux d’α-actinine lorsqu’il est phosphorylé. Également, il facilite l’hypertrophie des cardiomyocytes en présence de son activateur s’il est forcé hors du noyau ou en l’absence de son motif de liaison aux domaines PDZ. En conclusion, les résultats obtenus suggèrent que p38γ exerce bel et bien un rôle dans le maintien structural des cardiomyocytes par l’intermédiaire de l’α-actinine.

Doctor of Philosophy(PhD) / This thesis proposes the use of the augmented system form - a generalisation of the information form representing both observations and states. In conjunction with this, this thesis proposes a novel graph representation for the estimation problem together with a graph based linear direct solving algorithm. The augmented system form is a mathematical description of the estimation problem showing the states and observations. The augmented system form allows a more general range of factorisation orders among the observations and states, which is essential for constraints and is beneficial for sparsity and numerical reasons. The proposed graph structure is a novel sparse data structure providing more symmetric access and faster traversal and modification operations than the compressed-sparse-column (CSC) sparse matrix format. The graph structure was developed as a fundamental underlying structure for the formulation of sparse estimation problems. This graph-theoretic representation replaces conventional sparse matrix representations for the estimation states, observations and their interconnections. This thesis contributes a new implementation of the indefinite LDL factorisation algorithm based entirely in the graph structure. This direct solving algorithm was developed in order to exploit the above new approaches of this thesis. The factorisation operations consist of accessing adjacencies and modifying the graph edges. The developed solving algorithm demonstrates the significant differences in the form and approach of the graph-embedded algorithm compared to a conventional matrix implementation. The contributions proposed in this thesis improve estimation methods by providing novel mathematical data structures used to represent states, observations and the sparse links between them. These offer improved flexibility and capabilities which are exploited in the solving algorithm. The contributions constitute a new framework for the development of future online and incremental solving, data association and analysis algorithms for online, large scale localisation and mapping.

Offshore software development (OSD) is a leading business sector in the present ‘glocal’ IT marketplace, and vendors in different countries are opening software development centres worldwide to take advantage of the new business opportunities. However, software development is both a technical and a social process as various software modules need to be integrated, which requires ongoing interaction between the stakeholders. The software modules rely upon local knowledge regarding customer wants, project specific features, chosen design methodologies by development team members and synchronisation of activities to confirm the next design iteration. This study focuses on knowledge sharing processes involving the interplay between acquiring local knowledge and applying the knowledge acquired into the design of the client-specific software builds. New knowledge is created as new processes are applied and new outcomes realised, resulting in re-definition of software development practices. Building on existing theories with empirical case study evidence, this research reveals the socio-technical influences on knowledge management in the OSD process. Ideographic research methods have been applied to bring sensitivity in the everyday organisational activities for knowledge sharing across diverse social and cultural groups within two country contexts (New Zealand and India). Empirical data from ten case studies is used to inductively develop a conceptual framework, which has been applied to make within case and cross case comparisons across three levels of analysis (micro, meso and macro) for knowledge sharing. The micro level analysis explores individual key success drivers (behaviours and methodologies), the meso level explores organisational level practices (work processes and structures) and the macro level gives a holistic evaluation across two country contexts. Country contexts reveal that New Zealand vendors share closer cultural proximity with their clients, are engaged in client facing skills and have further outsourced software development tasks to other low cost countries. The Indian vendors are involved in software construction, prefer technical specialist skills and have defined more discipline in their software development processes. The thesis offers new insights on how vendors’ shape their software development styles based upon their beliefs and understanding of the offshore market and is especially relevant to both vendors and clients who intend venturing into the offshore market.

Offshore software development (OSD) is a leading business sector in the present ‘glocal’ IT marketplace, and vendors in different countries are opening software development centres worldwide to take advantage of the new business opportunities. However, software development is both a technical and a social process as various software modules need to be integrated, which requires ongoing interaction between the stakeholders. The software modules rely upon local knowledge regarding customer wants, project specific features, chosen design methodologies by development team members and synchronisation of activities to confirm the next design iteration. This study focuses on knowledge sharing processes involving the interplay between acquiring local knowledge and applying the knowledge acquired into the design of the client-specific software builds. New knowledge is created as new processes are applied and new outcomes realised, resulting in re-definition of software development practices. Building on existing theories with empirical case study evidence, this research reveals the socio-technical influences on knowledge management in the OSD process. Ideographic research methods have been applied to bring sensitivity in the everyday organisational activities for knowledge sharing across diverse social and cultural groups within two country contexts (New Zealand and India). Empirical data from ten case studies is used to inductively develop a conceptual framework, which has been applied to make within case and cross case comparisons across three levels of analysis (micro, meso and macro) for knowledge sharing. The micro level analysis explores individual key success drivers (behaviours and methodologies), the meso level explores organisational level practices (work processes and structures) and the macro level gives a holistic evaluation across two country contexts. Country contexts reveal that New Zealand vendors share closer cultural proximity with their clients, are engaged in client facing skills and have further outsourced software development tasks to other low cost countries. The Indian vendors are involved in software construction, prefer technical specialist skills and have defined more discipline in their software development processes. The thesis offers new insights on how vendors’ shape their software development styles based upon their beliefs and understanding of the offshore market and is especially relevant to both vendors and clients who intend venturing into the offshore market.

This thesis deals with the Simultaneous Localisation and Mapping algorithm as it pertains to the deployment of mobile systems in unknown environments. Simultaneous Localisation and Mapping (SLAM) as defined in this thesis is the process of concurrently building up a map of the environment and using this map to obtain improved estimates of the location of the vehicle. In essence, the vehicle relies on its ability to extract useful navigation information from the data returned by its sensors. The vehicle typically starts at an unknown location with no a priori knowledge of landmark locations. From relative observations of landmarks, it simultaneously computes an estimate of vehicle location and an estimate of landmark locations. While continuing in motion, the vehicle builds a complete map of landmarks and uses these to provide continuous estimates of the vehicle location. The potential for this type of navigation system for autonomous systems operating in unknown environments is enormous. One significant obstacle on the road to the implementation and deployment of large scale SLAM algorithms is the computational effort required to maintain the correlation information between features in the map and between the features and the vehicle. Performing the update of the covariance matrix is of O(n�) for a straightforward implementation of the Kalman Filter. In the case of the SLAM algorithm, this complexity can be reduced to O(n�) given the sparse nature of typical observations. Even so, this implies that the computational effort will grow with the square of the number of features maintained in the map. For maps containing more than a few tens of features, this computational burden will quickly make the update intractable - especially if the observation rates are high. An effective map-management technique is therefore required in order to help manage this complexity. The major contributions of this thesis arise from the formulation of a new approach to the mapping of terrain features that provides improved computational efficiency in the SLAM algorithm. Rather than incorporating every observation directly into the global map of the environment, the Constrained Local Submap Filter (CLSF) relies on creating an independent, local submap of the features in the immediate vicinity of the vehicle. This local submap is then periodically fused into the global map of the environment. This representation is shown to reduce the computational complexity of maintaining the global map estimates as well as improving the data association process by allowing the association decisions to be deferred until an improved local picture of the environment is available. This approach also lends itself well to three natural extensions to the representation that are also outlined in the thesis. These include the prospect of deploying multi-vehicle SLAM, the Constrained Relative Submap Filter and a novel feature initialisation technique. Results of this work are presented both in simulation and using real data collected during deployment of a submersible vehicle equipped with scanning sonar.

Doctor of Philosophy (PhD) / This thesis addresses the issues arising in range-only localisation and tracking using Radio Frequency Received Signal Strength Indicator measurements. One of the key issues in Radio Frequency (RF) based localisation and tracking applications is to obtain an accurate sensor representation. Such a sensor model is one of the prerequisites to achieve high accuracy and precision in the localisation and tracking task. The sensor models used at present for this task are very simplistic, and as a consequence are unable to achieve highly accurate and precise localisation. While such an accurate sensor description is desirable it has not been presented for RF sensors. This thesis addresses the task of obtaining an accurate sensor model for RF sensors. The major drawbacks of the most commonly used model, the nth power model, are demonstrated. A new model to satisfy the necessary requirements for high accuracy localisation is developed. This model is based on theoretical considerations and experimental data. It depicts the real occurring behaviour of RF sensors more closely than the models used so far for RF based range-only localisation. The use of this better sensor representation offers the possibility of achieving more accurate localisation. The expected performance of the alternative sensor model is compared to the commonly used nth power model. Furthermore, the inherent properties of the new sensor model are presented and their ramifications with regards to the goal of achieving highly accurate localisation are discussed. In addition to the sensor model development, the well-known probabilistic filtering techniques Kalman Filter, Particle Filter and Histogram Filter are compared and used to implement 1-dimensional and 2-dimensional range-only trackers. The filtering techniques are evaluated with respect to their suitability for appropriately handling the new multi-modal sensor model and the resulting multi-modal state distributions, and to provide correct and conclusive localisation and tracking results. Results from experiments using real data obtained in outdoor environments with a prototype RF localisation system as well as results obtained from simulations are presented in this thesis to validate the theoretical findings and the newly developed sensor model.

Doctor of Philosophy(PhD) / This thesis proposes the use of the augmented system form - a generalisation of the information form representing both observations and states. In conjunction with this, this thesis proposes a novel graph representation for the estimation problem together with a graph based linear direct solving algorithm. The augmented system form is a mathematical description of the estimation problem showing the states and observations. The augmented system form allows a more general range of factorisation orders among the observations and states, which is essential for constraints and is beneficial for sparsity and numerical reasons. The proposed graph structure is a novel sparse data structure providing more symmetric access and faster traversal and modification operations than the compressed-sparse-column (CSC) sparse matrix format. The graph structure was developed as a fundamental underlying structure for the formulation of sparse estimation problems. This graph-theoretic representation replaces conventional sparse matrix representations for the estimation states, observations and their interconnections. This thesis contributes a new implementation of the indefinite LDL factorisation algorithm based entirely in the graph structure. This direct solving algorithm was developed in order to exploit the above new approaches of this thesis. The factorisation operations consist of accessing adjacencies and modifying the graph edges. The developed solving algorithm demonstrates the significant differences in the form and approach of the graph-embedded algorithm compared to a conventional matrix implementation. The contributions proposed in this thesis improve estimation methods by providing novel mathematical data structures used to represent states, observations and the sparse links between them. These offer improved flexibility and capabilities which are exploited in the solving algorithm. The contributions constitute a new framework for the development of future online and incremental solving, data association and analysis algorithms for online, large scale localisation and mapping.