Optical flow, caused by relative motion of the object and the viewer, is the distribution of apparent velocities of brightness pattern in an image. The advantage of the optical-flow-based visual servo method is that feature of the object does not need to be defined or known in advance. This research plans to build an image servo technique to deal with the problem of 3D relative motion of the viewer and the environment. The images are treated as input and output signals of the control system and are fed back to extract the relative velocity information between contiguous image patterns. Then the video camera will automatically follow the motion to maintain the target image unchanged.

The digital image processing has been developed for a long time. The image detection and tracking are involved to a variety of digital techniques. In this research we introduce the digital image processing techniques, base on a boosted cascade of simple features to develop a face detection and tracking system. Due to a large amount of computation in face detection under the complex environment will affect the detection rate and velocity efficiency. Therefore, we use the extended feature and set of 45ﾟ rotated feature with fast feature computation which called the integral image, combine with the deformable templates. We can compute a part of the image block to reduce the computation and improve the system. In the PAN-TILT unit, we use fuzzy logic. The results of experiment show that system is robust and fast.

In this research, a single PAN-TILT image servo system has been developed with real-time face tracing technology. First, the target face is detected, and then the target template is kept at the image center with the integration of optical flow algorithm and control theory. In motion control, back-propagation neural network is taken to predict and estimate the target position. Experiments are made to analyze the performance of the video servo control system.

A monocular visual servo system for a target with variable shape has been developed in this paper. It consists of two parts: an image-processing unit and a servo control unit. For the image-processing unit, the motion between the target and image center is determined by a template match approach. The image is grabbed by the camera equipped on a Pan-Tilt robot and the robot is controlled to track the target by maintaining the target on the image center. However, the template needs to be updated when the target deforms. For the servo control unit, the movement is estimated by the Kalman filter technique to enhance the tracking performance of the visual servo system.

Target tracking is a broad subject area extensively studied in many engineering disciplines. In this thesis, target tracking implies the temporal estimation of target features such as the target's direction-of-arrival (DOA), the target's boundary pixels in a sequence of images, and/or the target's position in space. For multiple target tracking, we have introduced a new motion model that incorporates an acceleration component along the heading direction of the target. We have also shown that the target motion parameters can be considered part of a more general feature set for target tracking, e.g., target frequencies, which may be unrelated to the target motion, can be used to improve the tracking performance. We have introduced an acoustic multiple-target tracker using a flexible observation model based on an image tracking approach by assuming that the DOA observations might be spurious and that some of the DOAs might be missing in the observation set. We have also addressed the acoustic calibration problem from sources of opportunity such as beacons or a moving source. We have derived and compared several calibration methods for the case where the node can hear a moving source whose position can be reported back to the node. The particle filter, as a recursive algorithm, requires an initialization phase prior to tracking a state vector. The Metropolis-Hastings (MH) algorithm has been used for sampling from intractable multivariate target distributions and is well suited for the initialization problem. Since the particle filter only needs samples around the mode, we have modified the MH algorithm to generate samples distributed around the modes of the target posterior. By simulations, we show that this mode hungry algorithm converges an order of magnitude faster than the original MH scheme. Finally, we have developed a general framework for the joint state-space tracking problem. A proposal strategy for joint state-space tracking using the particle filters is defined by carefully placing the random support of the joint filter in the region where the final posterior is likely to lie. Computer simulations demonstrate improved performance and robustness of the joint state-space when using the new particle proposal strategy.

Cette thèse traite la problématique d'estimation de la pose (position relative et orientation) d'un corps rigide en mouvement dans l’espace 3D par fusion de données issues de capteurs inertiels et visuels. Les mesures inertielles sont fournies à partir d’une centrale inertielle composée de gyroscopes 3 axes et d’accéléromètres 3 axes. Les données visuelles sont issues d’une caméra. Celle-ci est positionnée sur le corps rigide en mouvement, elle fournit des images représentatives du champ visuel perçu. Ainsi, les mesures implicites des directions des lignes, supposées fixes dans l’espace de la scène, projetées sur le plan de l’image seront utilisées dans l’algorithme d’estimation de l’attitude. La démarche consistait d’abord à traiter le problème de la mesure issue du capteur visuel sur une longue séquence en utilisant les caractéristiques de l’image. Ainsi, un algorithme de suivi de lignes a été proposé en se basant sur les techniques de calcul du flux optique des points extraits des lignes à suivre et utilisant une approche de mise en correspondance par minimisation de la distance euclidienne. Par la suite, un observateur conçu dans l’espace SO(3) a été proposé afin d’estimer l’orientation relative du corps rigide dans la scène 3D en fusionnant les données issues de l’algorithme de suivi de lignes avec les données des gyroscopes. Le gain de l’observateur a été élaboré en utilisant un filtre de Kalman de type M.E.K.F. (Multiplicative Extended Kalman Filter). Le problème de l’ambigüité du signe dû à la mesure implicite des directions des lignes a été considéré dans la conception de cet observateur. Enfin, l’estimation de la position relative et de la vitesse absolue du corps rigide dans la scène 3D a été traitée. Deux observateurs ont été proposés : le premier est un observateur en cascade avec découplage entre l’estimation de l’attitude et l’estimation de la position. L’estimation issue de l’observateur d’attitude alimente un observateur non linéaire utilisant des mesures issues des accéléromètres afin de fournir une estimation de la position relative et de la vitesse absolue du corps rigide. Le deuxième observateur, conçu quant à lui directement dans SE(3) , utilise un filtre de Kalman de type M.E.K.F afin d’estimer la pose par fusion de données inertielles (accéléromètres, gyromètres) et des données visuelles. Les performances des méthodes proposées sont illustrées et validées par différents résultats de simulation / AbstractThis thesis addresses the problems of pose estimation of a rigid body moving in 3D space by fusing data from inertial and visual sensors. The inertial measurements are provided from an I.M.U. (Inertial Measurement Unit) composed by accelerometers and gyroscopes. Visual data are from cameras, which positioned on the moving object, provide images representative of the perceived visual field. Thus, the implicit measure directions of fixed lines in the space of the scene from their projections on the plane of the image will be used in the attitude estimation. The approach was first to address the problem of measuring visual sensors after a long sequence using the characteristics of the image. Thus, a line tracking algorithm has been proposed based on optical flow of the extracted points and line matching approach by minimizing the Euclidean distance. Thereafter, an observer in the SO(3) space has been proposed to estimate the relative orientation of the object in the 3D scene by merging the data from the proposed lines tracking algorithm with Gyro data. The observer gain was developed using a Kalman filter type M.E.K.F. (Multiplicative Extended Kalman Filter). The problem of ambiguity in the sign of the measurement directions of the lines was considered in the design of the observer. Finally, the estimation of the relative position and the absolute velocity of the rigid body in the 3D scene have been processed. Two observers were proposed: the first one is an observer cascaded with decoupled from the estimation of the attitude and position estimation. The estimation result of the attitude observer feeds a nonlinear observer using measurements from the accelerometers in order to provide an estimate of the relative position and the absolute velocity of the rigid body. The second observer, designed directly in SE (3) for simultaneously estimating the position and orientation of a rigid body in 3D scene by fusing inertial data (accelerometers, gyroscopes), and visual data using a Kalman filter (M.E.K.F.). The performance of the proposed methods are illustrated and validated by different simulation results

Soft soil reinforcement by inclusion is a growing technique caracterized by a pile grid and a granular embankment introduced between the reinforced soil and the structure. Unlike traditionnal methods, the load is partially transferred to the pile heads by arching in the embankment. The application area of this research focuses on the shallow foundations case, in which the thickness of the embankment is small. The litterature review shows that only a few studies were dedicated to that case, and that fundamental questions remains concerning the load transfer in the embankment. Chosen method for this research consists in two-dimensionnal physical modelling, analysis of the conducted simulations, and development of an analytical model in order to predict the load transfer to the piles by arching in the embankment. The results of this PhD thesis provide original elements of evidence of the load transfer in the studied system, proposes an analytical model based on block division of the granular embankment by shear bands - which is in good agreement with experimental data - and lead to a better understanding of arching in soils.

Soft soil reinforcement by inclusion is a growing technique caracterized by a pile grid and a granular embankment introduced between the reinforced soil and the structure. Unlike traditionnal methods, the load is partially transferred to the pile heads by arching in the embankment. The application area of this research focuses on the shallow foundations case, in which the thickness of the embankment is small. The litterature review shows that only a few studies were dedicated to that case, and that fundamental questions remains concerning the load transfer in the embankment. Chosen method for this research consists in two-dimensionnal physical modelling, analysis of the conducted simulations, and development of an analytical model in order to predict the load transfer to the piles by arching in the embankment. The results of this PhD thesis provide original elements of evidence of the load transfer in the studied system, proposes an analytical model based on block division of the granular embankment by shear bands - which is in good agreement with experimental data - and lead to a better understanding of arching in soils. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished