The aim of the Master’s Thesis work is to study and develop algorithms for autonomous travel of a UGV (Unmanned Ground Vehicle). A vehicle for the mounting of sensors has been constructed in order to perform the work. Since the UGV is to be used outdoor in urban areas, GPS can be used. To improve precision and robustness, inertial navigation is used in addition to GPS, since GPS reception is likely to be diminished in such areas. The sensors used for navigation are consequently GPS, magnetometers, accelerometers, gyroscopes, tachometers and ultra sonic sensors measuring distance to be used in detection of obstacles. The system has been implemented in Matlab. Two alternative methods of navigation with sensor fusion have been developed; one is a decentralized method with Kalman filtering using an error model and the other is a centralized particle filter using an all-embracing model of the vehicle. The two methods have been evaluated and compared. Test results show that the two methods perform equivalently. The autonomous travel is undertaken between predetermined waypoints. In order to steer the vehicle a PID-controller based on the error between heading and its reference value is used. The computation of the reference value is based on position and heading in comparison to the desired path. The system has been tested using different routes and the results show an evident improvement of the precision in navigation compared to using only GPS-data. This holds for both navigation methods. Simulation of collision avoidance using virtual force fields shows satisfying results as well as terrain navigation with coordinate map referencing. / Examensarbetet är en studie i utveckling av algoritmer för autonom förflyttning av en UGV (eng Unmanned Ground Vehicle). För ändamålet har en farkost konstruerats där budgetsensorer för navigering används. Farkosten är tänkt att färdas utomhus i tätbebyggt område och GPS används. För förbättring av noggrannhet och robusthet vid dålig GPS-mottagning används även sensorer för tröghetsnavigering vilket här innebär magnetometrar, accelerometrar, gyron och tachometrar. För hinderdetektering finns avståndsmätande ultraljudssonar. Systemet som tagits fram har implementerats i realtid i Matlab. Två olika navigeringsmetoder med sensorfusion har utprovats; en decentraliserad variant med kalmanfilter som är uppbyggd kring felmodeller och en centraliserad variant med ett partikelfilter som använder en helhetsmodell för farkosten. De båda navigeringsmetoderna har utvärderats och jämförts. Resultat visar att de båda metoderna presterar likvärdigt. Den autonoma förflyttningen utförs mellan förutbestämda brytpunkter. För att styra farkosten har en PID-regulator baserad på felet mellan kurs och börvärde använts. Börvärdet på kurs baseras på nuvarande position och riktning relativt den önskade färdvägen. Olika körsituationer har testats och resultaten visar en markant förbättring av navigeringsprecisionen jämfört med endast GPS-mätningar för både kalman- och partikelfilter. Simuleringar på vektorfältsstyrning med virtuella kraftfält för att undvika hinder har utförts med goda resultat liksom simuleringar av kartreferenspositionering.

The main topic of the paper is on-line filtering for non-Gaussian dynamic (state space) models by approximate computation of the first two posterior moments using efficient numerical integration. Based on approximating the prior of the state vector by a normal density, we prove that the posterior moments of the state vector are related to the posterior moments of the linear predictor in a simple way. For the linear predictor Gauss-Hermite integration is carried out with automatic reparametrization based on an approximate posterior mode filter. We illustrate how further topics in applied state space modelling such as estimating hyperparameters, computing model likelihoods and predictive residuals, are managed by integration-based Kalman-filtering. The methodology derived in the paper is applied to on-line monitoring of ecological time series and filtering for small count data. (author's abstract) / Series: Forschungsberichte / Institut für Statistik

The tremendous growth of Web has made information overload problem increasingly serious. Users are often confused by huge amount of information available on the internet and they are faced with the problem of finding the most relevant information that meets their needs. Recommender systems have proven to be an important solution approach to this problem. This thesis will present OPENMORE, a movie recommendation system, which is primarily based on content-based filtering technique. The distinctive point of this study lies in the methodology used to construct and update user and item profiles and the optimizations used to fine-tune the constructed user models. The proposed system arranges movie content data as features of a set of dimension slots, where each feature is assigned a stable feature weight regardless of individual movies. These feature weights and the explicit feedbacks provided by the user are then used to construct the user profile, which is fine-tuned through a set of optimization mechanisms. Users are enabled to view their profile, update them and create multiple contexts where they can provide negative and positive feedback for the movies on the feature level.

In this thesis we study Particle Filter for visual tracking applications. The sequential Monte Carlo methods or Particle Filter provides approximate solutions when the tracking problem involves non-linear and/or non-Gaussian state space models. Also in this study, in order to make the visual tracker robust against change in target appearance and unexpected target motion, an adaptive target appearance model and a first order motion estimator are embedded in particle filtering. Additionally, since pixels that don&rsquo / t belong to target makes the motion estimation biased, the algorithm includes robust estimators to make the tracker reliable. Within the scope of this thesis the visual tracker proposed in [5] is implemented and the same problem is solved by proposing a Rao-Blackwellized Particle Filter. To deal with problems encountered during the implementation phase of the algorithm some improvements are made such as utilizing learning rate for the computation of adaptive velocity estimation. Moreover, some precautions are taken such as checking the velocity estimations to validate them. Finally, we have done several experiments both in indoor and outdoor environments to demonstrate the effectiveness and robustness of the implemented algorithm. Experimental results show that most of the time the visual tracker performs well. On the other hand the drawbacks of the implemented tracker are indicated and we explain how to eliminate them.

Recently, it has become more and more difficult for the existing web based systems to locate or retrieve any kind of relevant information, due to the rapid growth of the World Wide Web (WWW) in terms of the information space and the amount of the users in that space. However, in today&#039 / s world, many systems and approaches make it possible for the users to be guided by the recommendations that they provide about new items such as articles, news, books, music, and movies. However, a lot of traditional recommender systems result in failure when the data to be used throughout the recommendation process is sparse. In another sense, when there exists an inadequate number of items or users in the system, unsuccessful recommendations are produced. Within this thesis work, ReMovender, a web based movie recommendation system, which uses a content boosted collaborative filtering approach, will be presented. ReMovender combines the local/global similarity and missing data prediction v techniques in order to handle the previously mentioned sparseness problem effectively. Besides, by putting the content information of the movies into consideration during the item similarity calculations, the goal of making more successful and realistic predictions is achieved.

This thesis tackles the visual tracking problem as a target contour estimation problem in the face of corrupted measurements. The major aim is to design robust recursive curve filters for accurate contour-based tracking. The state-space representation adopted comprises of a group component and a shape component describing the rigid motion and the non-rigid shape deformation respectively; filtering strategies on each component are then decoupled. The thesis considers two implicit curve descriptors, a classification probability field and the traditional signed distance function, and aims to develop an optimal probabilistic contour observer and locally optimal curve filters. For the former, introducing a novel probabilistic shape description simplifies the filtering problem on the infinite-dimensional space of closed curves to a series of point-wise filtering tasks. The definition and justification of a novel update model suited to the shape space, the derivation of the filtering equations and the relation to Kalman filtering are studied. In addition to the temporal consistency provided by the filtering, extensions involving distributed filtering methods are considered in order to maintain spatial consistency. For the latter, locally optimal closed curve filtering strategies involving curve velocities are explored. The introduction of a local, linear description for planar curve variation and curve uncertainty enables the derivation of a mechanism for estimating the optimal gain associated to the curve filtering process, given quantitative uncertainty levels. Experiments on synthetic and real sequences of images validate the filtering designs.

The thesis outlines the development and application of a few novel dynamic state estimation based methods for estimation of parameters of vibrating engineering structures. The study investigates strategies for data fusion from multiple tests of possibly different types and different sensor quantities through the introduction of a common pseudo-time parameter. These strategies have been developed within the framework of Kalman and particle filtering techniques. The proposed methods are applied to a suite of problems that includes laboratory and field studies with a primary focus on finite element model updating of bridge structures and vehicle structure interaction problems. The study also describes how finite element models residing in commercially available softwares can be made to communicate with database of measurements via a particle filtering algorithm developed on the Matlab platform. The thesis is divided into six chapters and an appendix. A review of literature on problems of structural system identification with emphasis on methods on dynamic state estimation techniques is presented in Chapter 1. The problem of system parameter idenfification when measurements originate from multiple tests and multiple sensors is considered in Chapter 2. and solution based on Neumann expansion of the structural static/dynamic stiffness matrix and Kalman filtering is proposed to tackle this problem. The question of decoupling the problem of parameter estimation from state estimation is also discussed. The avoidance of linearization of the stiffness matrix and solution of the parameter problems by using Monte Carlo filters is examined in Chapter 3. This also enables treatment of nonlinear structural mechanics problems. The proposed method is assessed using synthetic and laboratory measurement data. The problem of interfacing structural models residing in professional finite element analysis software with measured data via particle filtering algorithm developed on Matlab platform is considered in Chapter 4. Illustrative examples now cover laboratory studies on a beam structure and also filed studies on an existing multi-span masonry railway arch bridge. Identification of parameters of systems with strong nonlinearities, such, as a rectangular rubber sheet with a concentric hole, is also investigated. Studies on parameter identification in beam moving oscillator problem are reported in Chapter 5. The efficacy of particle filtering strategy in identifying parameters of this class of time varying system is demonstrated. A resume of contributions made and a few suggestions for further research are provided in Chapter 6. The appendix contains details of development of interfaces among finite element software(NISA), data base of measurements and particle filtering algorithm (developed on Matlab platform).

The problem of object detection deals with determining whether an instance of a given class of object is present or not. There are robust, supervised learning based algorithms available for object detection in an image. These image object detectors (image-based object detectors) use characteristics learnt from the training samples to find object and non-object regions. The characteristics used are such that the detectors work under a variety of conditions and hence are very robust. Object detection in video can be performed by using such a detector on each frame of the video sequence. This approach checks for presence of an object around each pixel, at different scales. Such a frame-based approach completely ignores the temporal continuity inherent in the video. The detector declares presence of the object independent of what has happened in the past frames. Also, various visual cues such as motion and color, which give hints about the location of the object, are not used. The current work is aimed at building a generic framework for using a supervised learning based image object detector for video that exploits temporal continuity and the presence of various visual cues. We use temporal continuity and visual cues to speed up the detection and improve detection accuracy by considering past detection results. We propose a generic framework, based on Experiential Sampling [1], which considers temporal continuity and visual cues to focus on a relevant subset of each frame. We determine some key positions in each frame, called attention samples, and object detection is performed only at scales with these positions as centers. These key positions are statistical samples from a density function that is estimated based on various visual cues, past experience and temporal continuity. This density estimation is modeled as a Bayesian Filtering problem and is carried out using Sequential Monte Carlo methods (also known as Particle Filtering), where a density is represented by a weighted sample set. The experiential sampling framework is inspired by Neisser’s perceptual cycle [2] and Itti-Koch’s static visual attention model[3]. In this work, we first use Basic Experiential Sampling as presented in[1]for object detection in video and show its limitations. To overcome these limitations, we extend the framework to effectively combine top-down and bottom-up visual attention phenomena. We use learning based detector’s response, which is a top-down cue, along with visual cues to improve attention estimate. To effectively handle multiple objects, we maintain a minimum number of attention samples per object. We propose to use motion as an alert cue to reduce the delay in detecting new objects entering the field of view. We use an inhibition map to avoid revisiting already attended regions. Finally, we improve detection accuracy by using a particle filter based detection scheme [4], also known as Track Before Detect (TBD). In this scheme, we compute likelihood of presence of the object based on current and past frame data. This likelihood is shown to be approximately equal to the product of average sample weights over past frames. Our framework results in a significant reduction in overall computation required by the object detector, with an improvement in accuracy while retaining its robustness. This enables the use of learning based image object detectors in real time video applications which otherwise are computationally expensive. We demonstrate the usefulness of this framework for frontal face detection in video. We use Viola-Jones’ frontal face detector[5] and color and motion visual cues. We show results for various cases such as sequences with single object, multiple objects, distracting background, moving camera, changing illumination, objects entering/exiting the frame, crossing objects, objects with pose variation and sequences with scene change. The main contributions of the thesis are i) We give an experiential sampling formulation for object detection in video. Many concepts like attention point and attention density which are vague in[1] are precisely defined. ii) We combine detector’s response along with visual cues to estimate attention. This is inspired by a combination of top-down and bottom-up attention maps in visual attention models. To the best of our knowledge, this is used for the first time for object detection in video. iii) In case of multiple objects, we highlight the problem with sample based density representation and solve by maintaining a minimum number of attention samples per object. iv) For objects first detected by the learning based detector, we propose to use a TBD scheme for their subsequent detections along with the learning based detector. This improves accuracy compared to using the learning based detector alone. This thesis is organized as follows . Chapter 1: In this chapter we present a brief survey of related work and define our problem. . Chapter 2: We present an overview of biological models that have motivated our work. . Chapter 3: We give the experiential sampling formulation as in previous work [1], show results and discuss its limitations. . Chapter 4: In this chapter, which is on Enhanced Experiential Sampling, we suggest enhancements to overcome limitations of basic experiential sampling. We propose track-before-detect scheme to improve detection accuracy. . Chapter 5: We conclude the thesis and give possible directions for future work in this area. . Appendix A: A description of video database used in this thesis. . Appendix B: A list of commonly used abbreviations and notations.

The osteocytes are the most abundant and longest living bone cells, embedded in the bone matrix. They are interconnected with each other through dendrites, located in slender canals called canaliculi. The osteocyte lacunae, cavities in which the cells are located, together with the canaliculi form a communication network throughout the bone matrix, permitting transport of nutrients, waste and signals. These cells were firstly considered passive, but lately it has become increasingly clear their role as mechanosensory cells and orchestrators of bone remodeling. Despite recent advances in imaging techniques, none of the available methods can provide an adequate 3D assessment of the lacuno-canalicular network (LCN). The aims of this thesis were to achieve 3D imaging of the LCN with synchrotron radiation X-ray computed tomography (SR-CT) and to develop tools for 3D detection and segmentation of this cell network, leading towards automatic quantification of this structure. We demonstrate the feasibility of parallel beam SR-CT to image in 3D the LCN (voxel~300 nm). This technique can provide data on both the morphology of the cell network and the composition of the bone matrix. Compared to the other 3D imaging methods, this enables imaging of tissue covering a number of cell lacunae three orders of magnitude greater, in a simpler and faster way. This makes possible the study of sets of specimens in order to reach biomedical conclusions. Furthermore, we propose the use of divergent holotomography, to image the ultrastructure of bone tissue (voxel~60 nm). The image reconstruction provides phase maps, obtained after the application of a suitable phase retrieval algorithm. This technique permits assessment of the cell network with higher accuracy and it enables the 3D organization of collagen fibres organization in the bone matrix, to be visualized for the first time. In order to obtain quantitative parameters on the geometry of the cell network, this has to be segmented. Due to the limitations in spatial resolution, canaliculi appear as 3D tube-like structures measuring only 1-3 voxels in diameter. This, combined with the noise, the low contrast and the large size of each image (8 GB), makes the segmentation a difficult task. We propose an image enhancement method, based on a 3D line filter combined with bilateral filtering. This enables improvement in canaliculi detection, reduction of the background noise and cell lacunae preservation. For the image segmentation we developed a method based on variational region growing. We propose two expressions for energy functionals to minimize in order to detect the desired structure, based on the 3D line filter map and the original image. Preliminary quantitative results on human femoral samples are obtained based on connected components analysis and a few observations related to the bone cell network and its relation with the bone matrix are presented.

Este trabalho trata da coordenação de robôs heterogêneos que consiste em um robô helicóptero e em múltiplos robôs móveis com rodas, de modo que estes sigam um líder. Para atingirmos este objetivo, neste trabalho desenvolve-se a estimação da atitude e posição de um robô utilizando filtros robustos. Os filros realizam a fusão dos sinais de uma unidade de medida inercial de baixo custo e de um receptor GPS (Global Positioning System) considerando um modelo de corpo rígido formulado em termos de quatérnios. Resultados experimentais são apresentados baseados em um estudo comparativo entre os filtros robustos e o filtro de Kalman. Desenvolve-se o controle de um robô móvel com rodas deslizantes e de um robô helicóptero. Para realizar o controle robusto utilzia-se um regulador robusto recursivo. Simulaçõoes do robô móvel com rodas deslizantes e do robô helicóptero são apresentadas utilizando os parâmetros de um robô móvel Pioneer 3AT e de um robô helicóptero RMAX. Desenvolve-se também um mini robô helicóptero e uma base de movimento para realização de testes de controle do mini robô helicóptero constiste de um mini helimodelo e de um piloto automático baseado em um microprocessador com conectividade Wi-Fi, uma unidade de medida inercial e uma placa de controle de servos motores. E por fim, desenvolve-se a coordenação de robôs heterogêneos de modo que estes sigam um líder. Para isto é utilziado um controlador descentralizado e cooperativo cuja finalidade é gerar trajetórias de referência para que os robôs heterogênos se movimentem em formação rígida. / This research deals with the coordination of the heterogeneous robots, consisting of a robotic helicopter and multiple uheeled mobile robots, to achieve this aim this work develops the estimation of attitude, heading and positions of a robot based on robust filters. The filters perform the fusion of the signals of a low-cost inertial measurement unit and a GPS (Global Positioning System) receiver considering a rigid body model formulated in terms of quaternions. Experimental results are presented based on a comparative study of the robust filters and Kalman filter. It develops a robust control scheme of a ski-steering mobile robot and a robotic helicopter. The robust control is performed through a the robust recursive regulator. Simulations of the skid-steering mobile robot and a robotic helicopter are presented using the parameters of a Pioneer 3AT and a robotic helicopter RMAX. It also develops a mini robotic helicopter and a motion base to perform control test of a mini robotic helicopter. The mini robotic helicopter consists of a mini helicopter TREX 450 XL and an autopilot based on a microprocessor with Wi-Fi, an inertial measurement unit a servo control board. To finish, it develops the coordination of the heterogeneous robots in order to perform a leader strategy based on the concept of rigid formation.