Extreme Learning Machine (ELM) is a training algorithm for Single-Layer Feed-forward Neural Network (SLFN). The difference in theory of ELM from other training algorithms is in the existence of explicitly-given solution due to the immutability of initialed weights. In practice, ELMs achieve performance similar to that of other state-of-the-art training techniques, while taking much less time to train a model. Experiments show that the speedup of training ELM is up to the 5 orders of magnitude comparing to standard Error Back-propagation algorithm. ELM is a recently discovered technique that has proved its efficiency in classic regression and classification tasks, including multi-class cases. In this thesis, extensions of ELMs for non-typical for Artificial Neural Networks (ANNs) problems are presented. The first extension, described in the third chapter, allows to use ELMs to get probabilistic outputs for multi-class classification problems. The standard way of solving this type of problems is based 'majority vote' of classifier's raw outputs. This approach can rise issues if the penalty for misclassification is different for different classes. In this case, having probability outputs would be more useful. In the scope of this extension, two methods are proposed. Additionally, an alternative way of interpreting probabilistic outputs is proposed. ELM method prove useful for non-linear dimensionality reduction and visualization, based on repetitive re-training and re-evaluation of model. The forth chapter introduces adaptations of ELM-based visualization for classification and regression tasks. A set of experiments has been conducted to prove that these adaptations provide better visualization results that can then be used for perform classification or regression on previously unseen samples. Shape registration of 3D models with non-isometric distortion is an open problem in 3D Computer Graphics and Computational Geometry. The fifth chapter discusses a novel approach for solving this problem by introducing a similarity metric for spectral descriptors. Practically, this approach has been implemented in two methods. The first one utilizes Siamese Neural Network to embed original spectral descriptors into a lower dimensional metric space, for which the Euclidean distance provides a good measure of similarity. The second method uses Extreme Learning Machines to learn similarity metric directly for original spectral descriptors. Over a set of experiments, the consistency of the proposed approach for solving deformable registration problem has been proven.

There is an increasing demand for robotics in dangerous and extreme conditions to limit human exposure and risk. An area in which robots are being considered as a support tool is in firefighting operations to reduce the number of firefighter injuries and deaths. One such application is to increase firefighting performance through localized fire suppression. This research focused on developing an autonomous suppression system for use on a mobile robotic platform. This included a real-time close proximity fire suppression approach, appropriate feature selection and probabilistic classification of water leaks and sprays, real-time trajectory estimation, and a feedback controller for error correction in longer-range firefighting. The close proximity suppression algorithm uses IR fire detection IR stereo processing to localize a fire. Feedback of the fire size and fire target was used to manipulate the nozzle for effective placement of the suppressant onto the fire and experimentally validated with tests in high and low visibility environments. To improve performance of autonomous suppression and for inspection tasks, identification of water sprays and leaks is a critical component. Bayesian classification was used to identify the features associated with water leaks and sprays in thermal images. Appropriate first and second order features were selected by using a multi-objective genetic algorithm optimization. Four textural features were selected as a method of discriminating water sprays and leaks from other non-water, high motion objects. Water classification was implemented into a real-time suppression system as a method of determining the yaw and pitch angle of a water nozzle. Estimation of the angle orientation provided an error estimate between the current path and desired nozzle orientation. A proportional-integral (PI) controller was used to correct for forced errors in fire targeting and performance and response was shown through indoor and outdoor suppression tests with wood-crib fires. The autonomous suppression algorithm was demonstrated through fire testing to be at least three times faster compared with suppression by an operator using tele-operation. / Ph. D.

Cette thèse est une contribution à la résolution du problème d’optimisation sous contrainte de fiabilité. Cette méthode de dimensionnement probabiliste vise à prendre en compte les incertitudes inhérentes au système à concevoir, en vue de proposer des solutions optimales et sûres. Le niveau de sûreté est quantifié par une probabilité de défaillance. Le problème d’optimisation consiste alors à s’assurer que cette probabilité reste inférieure à un seuil fixé par les donneurs d’ordres. La résolution de ce problème nécessite un grand nombre d’appels à la fonction d’état-limite caractérisant le problème de fiabilité sous-jacent. Ainsi,cette méthodologie devient complexe à appliquer dès lors que le dimensionnement s’appuie sur un modèle numérique coûteux à évaluer (e.g. un modèle aux éléments finis). Dans ce contexte, ce manuscrit propose une stratégie basée sur la substitution adaptative de la fonction d’état-limite par un méta-modèle par Krigeage. On s’est particulièrement employé à quantifier, réduire et finalement éliminer l’erreur commise par l’utilisation de ce méta-modèle en lieu et place du modèle original. La méthodologie proposée est appliquée au dimensionnement des coques géométriquement imparfaites soumises au flambement. / This thesis is a contribution to the resolution of the reliability-based design optimization problem. This probabilistic design approach is aimed at considering the uncertainty attached to the system of interest in order to provide optimal and safe solutions. The safety level is quantified in the form of a probability of failure. Then, the optimization problem consists in ensuring that this failure probability remains less than a threshold specified by the stakeholders. The resolution of this problem requires a high number of calls to the limit-state design function underlying the reliability analysis. Hence it becomes cumbersome when the limit-state function involves an expensive-to-evaluate numerical model (e.g. a finite element model). In this context, this manuscript proposes a surrogate-based strategy where the limit-state function is progressively replaced by a Kriging meta-model. A special interest has been given to quantifying, reducing and eventually eliminating the error introduced by the use of this meta-model instead of the original model. The proposed methodology is applied to the design of geometrically imperfect shells prone to buckling.