Dissertations / Theses on the topic 'SLAM'

PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO
COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR
FUNDAÇÃO DE APOIO À PESQUISA DO ESTADO DO RIO DE JANEIRO
PROGRAMA DE EXCELENCIA ACADEMICA
BOLSA NOTA 10
Atualmente, sistemas de visão computacional em computadores portáteis estão se tornando uma importante ferramenta de uso pessoal. Sistemas de visão para localização de objetos é uma área de pesquisa muito ativa. Essa dissertação propõe um algoritmo para localizar posições no espaço e objetos em ambientes não instrumentados com o uso de uma câmera web e um computador pessoal. Para isso, são usados dois algoritmos de rastreamento de marcadores para reinicializar frequentemente um algoritmo de Visual Simultaneous Localisation and Mapping. Essa dissertação também apresenta uma implementação e um conjunto de testes para validar o algoritmo proposto.
Nowadays, vision systems in portable computers are becoming an important tool for personal use. Vision systems for object localization are an active area of research. This dissertation proposes an algorithm to locate position and objects in a regular environment with the use of a simple webcam and a personal computer. To that end, we use two algorithms of marker tracking to reboot often a Visual Simultaneous Localisation and Mapping algorithm. This dissertation also presents an implementation and a set of tests that validate the proposed algorithm.

Thesis: Sc. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2016.
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 151-162).
The ability to construct an accurate model of the environment is an essential capability for mobile autonomous systems, enabling such fundamental functions as planning, navigation, and manipulation. However, the general form of this problem, simultaneous localization and mapping (SLAM), is typically formulated as a maximum-likelihood estimation (MLE) that requires solving a nonconvex nonlinear program, which is computationally hard. Current state-of-the-art SLAM algorithms address this difficulty by applying fast local optimization methods to compute a critical point of the MLE. While this approach has enabled significant advances in SLAM as a practical technology by admitting the development of fast and scalable estimation methods, it provides no guarantees on the quality of the recovered estimates. This lack of reliability in existing SLAM algorithms in turn presents a serious barrier to the development of robust autonomous systems generally. To address this problem, in this thesis we develop a suite of algorithms for SLAM that preserve the computational efficiency of existing state-of-the-art methods while additionally providing explicit performance guarantees. Our contribution is threefold. First, we develop a provably reliable method for performing fast local optimization in the online setting. Our algorithm, Robust Incremental least-Squares Estimation (RISE), maintains the superlinear convergence rate of existing state-of-the-art online SLAM solvers while providing superior robustness to nonlinearity and numerical ill-conditioning; in particular, we prove that RISE is globally convergent under very mild hypotheses (namely, that the objective is twice-continuously differentiable with bounded sublevel sets). We show experimentally that RISE's enhanced convergence properties lead to dramatically improved performance versus alternative methods on SLAM problems exhibiting strong nonlinearities, such as those encountered in visual mapping or when employing robust cost functions. Next, we address the lack of a priori performance guarantees when applying local optimization methods to the nonconvex SLAM MLE by proposing a post hoc verification method for computationally certifying the correctness of a recovered estimate for pose-graph SLAM. The crux of our approach is the development of a (convex) semidefinite relaxation of the SLAM MLE that is frequently exact in the low to moderate measurement noise regime. We show that when exactness holds, it is straightforward to construct an optimal solution Z* for this relaxation from an optimal solution X* of the SLAM problem; the dual solution Z* (whose optimality can be verified directly post hoc) then serves as a certificate of optimality for the solution X* from which it was constructed. Extensive evaluation on a variety of simulated and real-world pose-graph SLAM datasets shows that this verification method succeeds in certifying optimal solutions across a broad spectrum of operating conditions, including those typically encountered in application. Our final contribution is the development of SE-Sync, a pose-graph SLAM inference algorithm that employs a fast purpose-built optimization method to directly solve the aforementioned semidefinite relaxation, and thereby recover a certifiably globally optimal solution of the SLAM MLE whenever exactness holds. As in the case of our verification technique, extensive empirical evaluation on a variety of simulated and real-world datasets shows that SE-Sync is capable of recovering globally optimal pose-graph SLAM solutions across a broad spectrum of operating conditions (including those typically encountered in application), and does so at a computational cost that scales comparably with that of fast Newton-based local search techniques. Collectively, these algorithms provide fast and robust inference and verification methods for pose-graph SLAM in both the online and offline settings, and can be straightforwardly incorporated into existing concurrent filtering smoothing architectures. The result is a framework for real-time mapping and navigation that preserves the computational speed of current state-of-the-art techniques while delivering certifiably correct solutions.
by David M. Rosen.
Sc. D.

A core problem that must be solved by any practical visual SLAM system is the need to obtain correspondences throughout the image stream captured by a moving camera. Correspondences enable joint estimation of a moving camera's trajectory together with a 3D map of the observed scene. Visual SLAM pipelines commonly obtain correspondence through sparse feature matching techniques and construct maps using a composition of point, line or other simple primitives. The resulting sparse feature map representations provide sparsely furnished, incomplete reconstructions of the observed scene. Related techniques from multiple view stereo (MVS) achieve high quality dense reconstruction by obtaining dense correspondences over calibrated image sequences. Despite the usefulness of the resulting dense models, these techniques have been of limited use in visual SLAM systems. The computational complexity of estimating dense surface geometry has been a practical barrier to its use in real-time SLAM. Furthermore, MVS algorithms have typically required a fixed length, calibrated image sequence to be available throughout the optimisation --- a condition fundamentally at odds with the online nature of SLAM. With the availability of massively-parallel commodity computing hardware, we demonstrate new algorithms that achieve high quality incremental dense reconstruction within online visual SLAM. The result is a live dense reconstruction (LDR) of scenes that makes possible numerous applications that can utilise online surface modelling, for instance: planning robot interactions with unknown objects, augmented reality with characters that interact with the scene, or providing enhanced data for object recognition. The core of this thesis goes beyond LDR to demonstrate fully dense visual SLAM. We replace the sparse feature map representation with an incrementally updated, non-parametric, dense surface model. By enabling real-time dense depth map estimation through novel short baseline MVS, we can continuously update the scene model and further leverage its predictive capabilities to achieve robust camera pose estimation with direct whole image alignment. We demonstrate the capabilities of dense visual SLAM using a single moving passive camera, and also when real-time surface measurements are provided by a commodity depth camera. The results demonstrate state-of-the-art, pick-up-and-play 3D reconstruction and camera tracking systems useful in many real world scenarios.

Simultaneous Localisation and Mapping (SLAM) began as a technique to enable real-time robotic navigation on previously unexplored environments. The created maps however were designed for the sole purpose of localising the robot (i.e. what is the position and orientation of the robot in relation to the map) and several systems demonstrated the increasing descriptive power of map representations, which on vision-only SLAM solutions consisted of simple sparse corner-like features as well as edges, planes and most recently fully dense surfaces that abandon the notion of sparse structures altogether. Early sparse representations enjoyed the benefit of being simple to maintain as features could be added, optimised and removed independently while being memory and compute efficient, making them suitable for robust real-time camera tracking that relies on a consistent map. However, sparse representations are limiting when it comes to interaction, as for example, a robot aiming to safely navigate in an environment would need to sense complete surfaces in addition to empty space. Furthermore, sparse features can only be detected on highly-textured areas and during slow motion. Recent dense methods overcome the limitations of sparse methods as they can work in situations where corner features would fail to be detected due to blurry images created during rapid camera motion and also enable to correctly reason about occlusions and complete 3D surfaces, thus raising the interaction capabilities to new levels. This is only possible thanks to the advent of commodity parallel processing power and large amount of memory on Graphic Processing Units (GPUs) that needs careful consideration during algorithm design. However, increasing the map density makes creating consistent structures more challenging due to the vast amount of parameters to optimise and the interdependencies amongst them. More importantly, our interest is in making interaction even more sophisticated by abandoning the idea that an environment is a dense monolithic structure in favour of one composed of discrete detachable objects and bounded regions having physical properties and metadata. This work explores the development of a new type of visual SLAM system representing the map with semantically meaningful objects and planar regions which we call Dense Semantic SLAM, enabling new types of interaction where applications that can go beyond asking the question of "where am I" towards "what is around me and what can I do with it". In a way it can be seen as a return to lightweight sparse-based representations while keeping the predictive power of dense methods with added scene understanding at the object and region levels.

Localização e Mapeamento Simultâneos (SLAM) é uma das tarefas mais difíceis em robótica móvel, uma vez que existe uma dependência mútua entre a estimativa da localização do robô e a construção do mapa de ambiente. As estratégias de SLAM mais bem sucedidas focam na construção de um mapa métrico probabilístico empregando técnicas de filtragem Bayesiana. Embora tais métodos permitam a construção de soluções localmente consistentes e coerentes, o SLAM continua sendo um problema crítico em operações em ambientes grandes. Para contornar esta limitação, muitas estratégias dividem o ambiente em pequenas regiões, e formulam o problema de SLAM como uma combinação de múltiplos submapas métricos precisos associados em um mapa topológico. Este trabalho propõe um método de SLAM baseado nos algoritmos DP-SLAM (Distributed Particle SLAM) e SegSlam (Segmented SLAM). SegSLAM é um algoritmo que cria múltiplos submapas para cada região do ambiente, e posteriormente constrói o mapa global selecionando combinações de submapas. Por sua vez, DP-SLAM é um algoritmo de filtro de particulas Rao-Blackwellized que utiliza uma representação distribuída eficiente dos mapas das partículas, juntamente com a árvore de ascendência das partículas. A característica distribuída destas estruturas é favorável para a combinação de diferentes segmentos de mapa localmente precisos, o que aumenta a diversidade de soluções. O algoritmo proposto nesta dissertação, chamado SDP-SLAM, segmenta e combina diferentes hipóteses de trajetórias do robô, a fim de reconstruir o mapa do ambiente. Nossas principais contribuições são o desenvolvimento de novas estratégias para o casamento de submapas e para a estimativa de boas combinações de submapas. O SDP-SLAM foi avaliado através de experimentos realizados por um robô móvel operando em ambientes reais e simulados.
Simultaneous Localization and Mapping (SLAM) is one of the most difficult tasks in mobile robotics, since there is a mutual dependency between the estimation of the robot pose and the construction of the environment map. Most successful strategies in SLAM focus in building a probabilistic metric map employing Bayesian filtering techniques. While these methods allow the construction of consistent and coherent local solutions, the SLAM remains a critical problem in operations within large environments. To circumvent this limitation, many strategies divide the environment in small regions, and formulate the SLAM problem as a combination of multiple precise metric submaps associated in a topological map. This work proposes a SLAM method based on the Distributed Particle SLAM (DPSLAM) and the Segmented SLAM (SegSLAM) algorithms. SegSLAM is an algorithm that generates multiple submaps for every region of the environment, and then build the global map by selecting combinations of submaps. DP-SLAM is a Rao-Blackwellized particle filter algorithm that uses an efficient distributed representation of the particles maps associated with an ancestry tree of the particles. The distributed characteristic of these structures favors the combination of locally accurate map segments, that can increase the diversity of global level solutions. The algorithm proposed in this dissertation, called SDP-SLAM, segments and combines different hypotheses of robot trajectories to reconstruct the environment map. Our main contributions are the development of novel strategies for the matching of submaps and for the estimation of good submaps combinations. SDP-SLAM was evaluated through experiments performed by a mobile robot operating in real and simulated environments.

This paper explores the influences, ideas and motivations behind my MFA thesis exhibition. It primarily focuses on how I developed my work for the show in connection to my previous work as well as work created by other artists who explored the impacts of new media in the last decade. With the advancement of social media, digital technologies no longer have their infamous coldness. Our perceptions and the metaphors in language are all reflected onto the machines we create while in return they also shape and redefine our lives. It becomes increasingly difficult to talk about dialectics such as machine-human, virtual-real, and nature-culture. With the aid of some humor, I attempted to reflect on the marriage of these old oppositions and this paper will discuss the foundations of these ideas as well as my practice in general.

SLAM (Simulatenous Localization and Mapping) es quizá el problema más importante a solucionar en robótica para construir robots móviles verdaderamente autónomos. El SLAM es acerca de cómo un robot móvil opera en un entorno a priori desconocido, utilizando únicamente sus sensores de abordo, mientras construye un mapa de dicho entorno que al mismo tiempo utiliza para localizarse. Los sensores del robot tienen un gran impacto en los algoritmos usados en SLAM. Los primeros enfoques se centraron en el uso de sensores de rango como sonares o láseres. Sin embargo hay algunas desventajas relacionadas con su utilización: La asociación de datos es difícil, son costosos, habitualmente están limitados a mapas 2D y tienen alto costo computacional debido al gran número de características (features) que producen. Lo anterior ha propiciado que enfoques recientes se estén moviendo hacia el uso de cámaras como sensor principal. Estas se han vuelto muy atractivas para los investigadores de la robótica, dado que generan mucha información, facilitan la asociación de datos, están bien adaptadas para sistemas embebidos: son ligeras, baratas y ahorran energía. Usando visión, un robot puede localizarse así mismo usando objetos comunes como landmarks. Sin embargo, a diferencia de los sensores de rango, que proveen información angular y de rango, una cámara es un sensor proyectivo que mide el bearing (ángulo) respecto a objetos de la imagen. Por lo que la profundidad (range) no puede ser obtenida en una sola toma. Este hecho ha motivado la aparición de una nueva familia de métodos de SLAM: Los Bearing-Only SLAM methods, los cuales están basados en técnicas especiales para la inicialización de features, permitiendo el uso de sensores de bearing en SLAM. Esta tesis se centra en el estudio de la problemática del Bearing-Only SLAM: da una descripción extensa del tema, recapitula los retos actuales a resolver y propone nuevos métodos y algoritmos enfocados a tratar diferentes sub problemas concernientes esta problemática en general. Estos sub problemas deben de ser tratados, de manera que sea posible construir sistemas capaces de operar en entornos diversos y complejos. La investigación descrita en esta disertación ha sido dividida en tres partes: 3DOF Bearing-Only SLAM: El proceso de inicialización de nuevas features es quizá el sub problema más importante a tratar en Bearing-Only SLAM. En esta parte de la tesis se introduce un nuevo método llamado Delayed Inverse Depth Features Initialization (para 3DOF y asumiendo odometría). Este método utiliza una parametrización inversa, donde la profundidad e incertidumbre iníciales de cada feature son dinámicamente estimadas previamente a que una feature sea declarada como un nuevo landmark en el mapa estocástico. También se presenta un sistema de SLAM basado en sonido, llamado SSLAM el cual usa fuentes de sonido como features del mapa. La contribución del SSLAM es demostrar la viabilidad de la inclusión del sentido auditivo en SLAM y mostrar que es factible utilizar sensores alternativos en Bearing-Only SLAM. Métodos de asociación de datos para SLAM basado en visión: El problema de la asociación de datos es quizá uno de los problemas más difíciles en robótica y también uno de los sub problemas más importantes a tratar en SLAM. Consiste en determinar si las mediciones de un sensor tomadas en tiempos diferentes, corresponden al mismo objeto físico del mundo. En esta parte de la tesis, se proponen diferentes métodos que tratan el problema de la asociación de datos en un contexto de SLAM basado en visión. SLAM monocular de 6DOF: El SLAM monocular de 6DOF quizá representa la variante más extrema del SLAM, dado que una cámara en mano es utilizada como la única entrada sensorial del sistema. En esta parte de la tesis, se extiende el algoritmo de 2DOF Bearing-Only SLAM para ser aplicado en un contexto de SLAM monocular. También se propone un nuevo esquema llamado SLAM Monocular Distribuido, enfocado en el problema de construir y mantener mapas consistentes de grandes entornos en tiempo real. La idea es dividir la estimación total del sistema en dos procesos de estimación concurrentes. Primero un método actual de SLAM monocular (Virtual Sensor) es modificado como un complejo sensor virtual que emula sensores típicos, como el laser para medición de rango y encoders para odometría. Después otro método tradicional de SLAM (Global SLAM) es acoplado para construir y mantener el mapa final. Numerosas referencias bibliográficas, graficas, comparaciones, simulaciones y experimentos con datos reales de sensores, son presentador con el fin de mostrar el desempeño de los métodos propuestos.
Simultaneous Localization and Mapping (SLAM) is perhaps the most fundamental problem to solve in robotics in order to build truly autonomous mobile robots. SLAM is about on how can a mobile robot operate in an a priori unknown environment and use only onboard sensors to simultaneously build a map of its surroundings and use it to track its position. The robot’s sensors have a large impact on the algorithm used for SLAM. Early SLAM approaches focused on the use of range sensors as sonar rings or lasers. Nevertheless there are some disadvantages with the use of range sensors in SLAM: Correspondence or data association is difficult. They are expensive. They are generally limited to 2D maps and computational overhead due to large number of features. The aforementioned issues have propitiated that recent work is moving towards the use of cameras as the primary sensing modality. Cameras have become more and more interesting for the robotic research community, because it yield a lot of information allowing reliable data association. Cameras are well adapted for embedded systems: they are light, cheap and power saving. Using vision, a robot can localize itself using common objects as landmarks. On the other hand, at difference of range sensors (i.e. sonar or laser) which provides range and angular information, a camera is a projective sensor which measures the bearing of images features. Therefore depth information (range) cannot be obtained in a single frame. This fact has propitiated the emergence of a new family of SLAM methods: The Bearing-Only SLAM methods, which mainly relies in especial techniques for features system-initialization in order to enable the use of bearing sensors (as cameras) in SLAM systems. This thesis is focused on the study of the Bearing-Only SLAM problematic: It gives an extensive overview of the subject. It point out the principal challenges nowadays. And it presents new methods and algorithms which address different sub problems concerning to the Bearing-Only SLAM problematic. These sub problems must be solved, in order to build systems capable of operating in extremely diverse and complex environments. The research described in this dissertation has been divided into three parts: 3DOF Bearing-Only SLAM: The initialization process for new features is perhaps the most important sub problem for addressing in Bearing-Only SLAM. In this part of the thesis we introduce a novel method called Delayed Inverse Depth Features Initialization for a 3DOF odometry-available context. In this method, which uses an inverse depth parameterization, initial depth and uncertainty of each feature are dynamically estimated priors to add the new landmark in the stochastic map. We also present a Sound-based SLAM system, called SSLAM, which uses “Sound Sources” as map’s features. The main contributions of the SSLAM are demonstrating the viability on the inclusion of the hearing sense in SLAM and show that is straightforward to use alternative bearing in SLAM systems. Data association methods for camera-based SLAM: the data association problem is possibly one of the hardest problems in robotic and also one of the most important sub problems to solve in SLAM. The correspondence problem is the problem of determining if sensor measurements taken at different points in time correspond to the same physical object in the world. In this part of the thesis, we propose different methods for addressing the data association problem in a context of vision-based SLAM. 6DOF Monocular SLAM: 6-DOF monocular SLAM possibly represents the harder variant of SLAM, since a low cost hand-held camera is used as the only sensory input to the system. In this part of the thesis, we extend our 2DOF Bearing-Only SLAM algorithm for being used in a monocular SLAM context. Also a novel framework called Distributed Monocular SLAM is proposed for addressing the problem of building and maintaining a global and consistent map of large environments at real time. The key idea is to divide the whole estimation into two concurrent estimation processes. First a state of the art monocular SLAM method (Called Virtual Sensor) is modified as a complex virtual sensor that emulates typical sensors such as laser for range measurement and encoders for dead reckoning. Afterward, a classic SLAM method (called Global SLAM) is plugged in for building and maintaining the final map. Several references, graphics, comparisons, simulations and experiments with real data are presented in order to demonstrate the performance of the methods.

Visual Simultaneous Localisation and Mapping (SLAM) is concerned with simultaneously estimating the pose of a camera and a map of the environment from a sequence of images. Traditionally, sparse maps comprising isolated point features have been employed, which facilitate robust localisation but are not well suited to advanced applications. In this thesis, we present map representations that allow a more dense description of the environment. In one approach, planar features are used to represent textured planar surfaces in the scene. This model is applied within a visual SLAM framework based on the Extended Kalman Filter. We presents solutions to several challenges which arise from this approach.

A classic Computer Vision task is the estimation of a 3D map from a collection of images. This thesis explores the online simultaneous estimation of camera poses and map points, often called Visual Simultaneous Localisation and Mapping [VSLAM]. In the near future the use of visual information by autonomous cars is likely, since driving is a vision dominated process. For example, VSLAM could be used to estimate the position of the car in relation to objects of interest, such as the road, other cars and pedestrians. Aimed at the creation of a real-time, robust, loop closing, single camera SLAM system, the properties of several state-of-the-art VSLAM systems and related techniques are studied. The system goals cover several important, if difficult, problems, which makes a solution widely applicable. This thesis makes two contributions: A rigorous qualitative analysis of VSLAM methods and a system designed accordingly. A novel tracking by matching scheme is proposed, which, unlike the trackers used by many similar systems, is able to deal better with forward camera motion. The system estimates general motion with loop closure in real time. The system is compared to a state-of-the-art monocular VSLAM algorithm and found to be similar in speed and performance.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2017.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 161-173).
This thesis solves the classical problem of simultaneous localization and mapping (SLAM) in a fashion which avoids linearized approximations altogether. Based on creating virtual synthetic measurements, the algorithm uses a linear time-varying (LTV) Kalman observer, bypassing errors and approximations brought by the linearization process in traditional extended Kalman filtering (EKF) SLAM. Convergence rates of the algorithm are established using contraction analysis. Different combinations of sensor information can be exploited, such as bearing measurements, range measurements, optical flow, or time-to-contact. As illustrated in simulations, the proposed algorithm can solve SLAM problems in both 2D and 3D scenarios with guaranteed convergence rates in a full nonlinear context. A novel distributed algorithm SLAM-DUNK is proposed in the thesis. The algorithm uses virtual vehicles to achieve information exclusively from corresponding landmarks. Computation complexity is reduced to 0(n), with simulations on Victoria Park dataset to support the validity of the algorithm. In the final section of the thesis, we propose a general framework for cooperative navigation and mapping. The frameworks developed for three different use cases use the null space terms of SLAM problem to guarantee that robots starting with unknown initial conditions could converge to a shared consensus coordinate system with estimates reflecting the truth.
by Feng Tan.
Ph. D.

This thesis explores approaches to two problems in the frame-rate computation of a priori unknown 3D scene structure and camera pose using a single camera, or monocular simultaneous localisation and mapping. The thesis reflects two trends in vision in general and structure from motion in particular: (i) the move from directly recovered and towards learnt geometry; and (ii) the sparsification of otherwise dense direct methods. The first contributions mitigate scale drift. Beyond the inevitable accumulation of random error, monocular SLAM accumulates error via the depth/speed scaling ambiguity. Three solutions are investigated. The first detects objects of known class and size using fixed descriptors, and incorporates their measurements in the 3D map. Experiments using databases with ground truth show that metric accuracy can be restored over kilometre distances; and similar gains are made using a hand-held camera. Our second method avoids explicit feature choice, instead employing a deep convolutional neural network to yield depth priors. Relative depths are learnt well, but absolute depths less so, and recourse to database-wide scaling is investigated. The third approach uses a novel trained network to infer speed from imagery. The second part of the thesis develops sparsified direct methods for monocular SLAM. The first contribution is a novel camera tracker operating directly using affine image warping, but on patches around sparse corners. Camera pose is recovered with an accuracy at least equal to the state of the art, while requiring only half the computational time. The second introduces a least squares adjustment to sparsified direct map refinement, again using patches from sparse corners. The accuracy of its 3D structure estimation is compared with that from the widely used method of depth filtering. It is found empirically that the new method's accuracy is often higher than that of its filtering counterpart, but that the method is more troubled by occlusion.

Existing work in the field of monocular Simultaneous Localisation and Mapping (SLAM) has largely centred around sparse feature-based representations of the world. By tracking salient image patches across many frames of video, both the positions of the features and the motion of the camera can be inferred live. Within the visual SLAM community, there has been a focus on both increasing the number of features that can be tracked across an image and efficiently managing and adjusting this map of features in order to improve camera trajectory and feature location accuracy. Although prior research has looked at augmenting this map with more sophisticated features such as edgelets or planar patches, no incremental real-time system has yet made use of every pixel in the image to maximise camera trajectory estimation accuracy. Moreover, across many practical domains, these feature-based representations of the world fall short. In robotics, sparse feature-based models do not allow a robot to reason about free space and are not so useful for interaction. In augmented reality, sparse models do not allow us to place virtual objects behind real-ones and cannot enable virtual characters to interact with real objects. In this research we show how a dense surface model offers many advantages and we explore different methods of reasoning about dense surfaces over a sparse feature-based map. We continue by developing different methods for dense tracking and constrained dense SLAM in different applications such as spherical mosaicing. Finally, we show how live dense tracking can be tightly integrated with dense reconstruction to create a 6 DOF monocular live dense SLAM system which outperforms the current state of the art in many respects.

Pour assurer la navigation autonome d’un robot mobile, les traitements effectués pour sa localisation doivent être faits en ligne et doivent garantir une précision suffisante pour permettre au robot d’effectuer des tâches de haut niveau pour la navigation et l’évitement d’obstacles. Les auteurs de travaux basés sur le SLAM visuel (Simultaneous Localization And Mapping) tentent depuis quelques années de garantir le meilleur compromis rapidité/précision. La majorité des solutions SLAM visuel existantes sont basées sur une représentation éparse de l’environnement. En suivant des primitives visuelles sur plusieurs images, il est possible d’estimer la position 3D de ces primitives ainsi que les poses de la caméra. La communauté du SLAM visuel a concentré ses efforts sur l’augmentation du nombre de primitives visuelles suivies et sur l’ajustement de la carte 3D, afin d’améliorer l’estimation de la trajectoire de la caméra et les positions 3D des primitives. Cependant, la localisation par SLAM visuel présente souvent des dérives dues au cumul d’erreurs, et dans le cas du SLAM visuel monoculaire, la position de la caméra n’est connue qu’à un facteur d’échelle près. Ce dernier peut être fixé initialement mais dérive au cours du temps. Pour faire face à ces limitations, nous avons centré nos travaux de thèse sur la problématique suivante : intégrer des informations supplémentaires dans un algorithme de SLAM visuel monoculaire afin de mieux contraindre la trajectoire de la caméra et la reconstruction 3D. Ces contraintes ne doivent pas détériorer les performances calculatoires de l’algorithme initial et leur absence ne doit pas mettre l’algorithme en échec. C’est pour cela que nous avons choisi d’intégrer l’information de profondeur fournie par un capteur 3D (e.g. Microsoft Kinect) et des informations géométriques sur la structure de la scène. La première contribution de cette thèse est de modifier l’algorithme SLAM visuel monoculaire proposé par Mouragnon et al. (2006b) pour prendre en compte la mesure de profondeur fournie par un capteur 3D, en proposant particulièrement un ajustement de faisceaux qui combine, d’une manière simple, des informations visuelles et des informations de profondeur. La deuxième contribution est de proposer une nouvelle fonction de coût du même ajustement de faisceaux qui intègre, en plus des contraintes sur les profondeurs des points, des contraintes géométriques d’appartenance aux plans de la scène. Les solutions proposées ont été validées sur des séquences de synthèse et sur des séquences réelles, représentant des environnements variés. Ces solutions ont été comparées aux récentes méthodes de l’état de l’art. Les résultats obtenus montrent que les différentes contraintes développées permettent d’améliorer significativement la précision de la localisation du SLAM. De plus les solutions proposées sont faciles à déployer et peu couteuses en temps de calcul
To guarantee autonomous and safely navigation for a mobile robot, the processing achieved for its localization must be fast and accurate enough to enable the robot to perform high-level tasks for navigation and obstacle avoidance. The authors of Simultaneous Localization And Mapping (SLAM) based works, are trying since year, to ensure the speed/accuracy trade-off. Most existing works in the field of monocular (SLAM) has largely centered around sparse feature-based representations of the environment. By tracking salient image points across many frames of video, both the positions of the features and the motion of the camera can be inferred live. Within the visual SLAM community, there has been a focus on both increasing the number of features that can be tracked across an image and efficiently managing and adjusting this map of features in order to improve camera trajectory and feature location accuracy. However, visual SLAM suffers from some limitations. Indeed, with a single camera and without any assumptions or prior knowledge about the camera environment, rotation can be retrieved, but the translation is up to scale. Furthermore, visual monocular SLAM is an incremental process prone to small drifts in both pose measurement and scale, which when integrated over time, become increasingly significant over large distances. To cope with these limitations, we have centered our work around the following issues : integrate additional information into an existing monocular visual SLAM system, in order to constrain the camera localization and the mapping points. Provided that the high speed of the initial SLAM process is kept and the lack of these added constraints should not give rise to the failure of the process. For these last reasons, we have chosen to integrate the depth information provided by a 3D sensor (e.g. Microsoft Kinect) and geometric information about scene structure. The primary contribution of this work consists of modifying the SLAM algorithm proposed by Mouragnon et al. (2006b) to take into account the depth measurement provided by a 3D sensor. This consists of several rather straightforward changes, but also on a way to combine the depth and visual data in the bundle adjustment process. The second contribution is to propose a solution that uses, in addition to the depth and visual data, the constraints lying on points belonging to the plans of the scene. The proposed solutions have been validated on a synthetic sequences as well as on a real sequences, which depict various environments. These solutions have been compared to the state of art methods. The performances obtained with the previous solutions demonstrate that the additional constraints developed, improves significantly the accuracy and the robustness of the SLAM localization. Furthermore, these solutions are easy to roll out and not much time consuming

"The use of cameras and computer vision algorithms to provide state estimation for robotic systems has become increasingly popular, particularly for small mobile robots and unmanned aerial vehicles (UAVs). These algorithms extract information from the camera images and perform simultaneous localization and mapping (SLAM) to provide state estimation for path planning, obstacle avoidance, or 3D reconstruction of the environment. High resolution cameras have become inexpensive and are a lightweight and smaller alternative to laser scanners. UAVs often have monocular camera or stereo camera setups since payload and size impose the greatest restrictions on their flight time and maneuverability. This thesis explores ORB-SLAM, a popular Visual SLAM method that is appropriate for UAVs. Visual SLAM is computationally expensive and normally offloaded to computers in research environments. However, large UAVs with greater payload capacity may carry the necessary hardware for performing the algorithms. The inclusion of general-purpose GPUs on many of the newer single board computers allows for the potential of GPU-accelerated computation within a small board profile. For this reason, an NVidia Jetson board containing an NVidia Pascal GPU was used. CUDA, NVidia’s parallel computing platform, was used to accelerate monocular ORB-SLAM, achieving onboard Visual SLAM on a small UAV. Committee members:"

This master thesis is about modifying a certain monocular visual SLAM algorithm to address some of its limitations. The SLAM algorithm is not robust to quick camera motions and input images in which there are few visible features. A second camera and an inertial measurement unit was added to the hardware. Then a method for selecting the appropriate camera for tracking depending on the estimated number of features was implemented to solve the issue of few features. Experiments and results show that this method works well for slow motions. A gyrometer threshold method along with a motion model to solve the issue of quick motions was implemented and reviewed in this thesis.
Detta examensarbete handlar om att ta itu med några begränsningar som en viss monokulär visuell SLAM-algoritm har. SLAM-algoritmen är inte robust mot snabba kamerarörelser och indatabilder som innehåller få karaktärsdrag. Genom att introducera en extra kamera, en accelerometer och en gyrometer, behandlas dessa problem i denna rapport. En metod för att välja kamera att hämta indatabilder från, baserat på det skattade antalet karaktärsdrag i respektive bild implementerades. Denna metod är tänkt att lösa problemet då indatabilder har få karaktärsdrag. Experiment visar att denna metod fungerar bra för långsamma rörelser. En metod som jämför gyrometerdata med ett tröskelvärde tillsammans med en rörelsemodell implementerades för att lösa problemen vid snabb rörelse. Dessa metoder undersöks och diskuteras i rapporten.

In mobile robotics, visual perception in unknown environments consists mainly of two tasks: generation of a map and localization of the agent within it, using images as input. This problem is commonly called Visual Simultaneous Localization and Mapping ( Visual SLA M). This project aims of increasing the capacities of spatial reasoning in robotics systems based on cognitive vision. Baycsian formulation of visual mapping is extended to consider geometric properties for increasing of scene understand- ing. In this search, this work presents a framework that covers the computa- tional and algorithmic perspectives in Cognitive Vision. proposed by Marr. A part of the work is devoted to develop the representational framework for describing the visual phenomenon in monocular SLAM, unifying concepts about the existent parametrization and extending the analysis to geometric properties and relations in the scene. This yields a novel strategy for aug- mented mapping with high lcvellandmarks based on planar surfaces. After, we explore the computational level of visual SLAM through a sym- bolic framework for describing the elements required for spatial reasoning, involved in visual SLAM. An ontology for spatial reasoning is proposed upon visual SLAM context. An axiomatic set in visual scenario is related to prop- erties in projective geometry, assuring a scheme for semantic attachment. Finally, a novel approach to solve visual SLAM based on spatial reasoning is presented, focusing on the algorithmic level. The model of latent geomet- ric constraints is presented as a non-parametric Baycsian extension of visual SLAM. The generative model produces multiples scenarios with different visual conditions. Although this thesis is focused on solving visual SLAM, the proposed ap- proach can be conceived as a methodology for the design of cognitive dynamic systems.

Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2013.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 91-94).
We show how to exploit temporal and spatial coherence of image observations to achieve efficient and effective text detection and decoding for a sensor suite moving through an environment rich in text at a variety of scales and orientations with respect to the observer. We use simultaneous localization and mapping (SLAM) to isolate planar "tiles" representing scene surfaces and prioritize each tile according to its distance and obliquity with respect to the sensor, and how recently (if ever) and at what scale the tile has been inspected for text. We can also incorporate prior expectations about the spatial locus and scale at which text occurs in the world, for example more often on vertical surfaces than non-vertical surfaces, and more often at shoulder height than at knee height. Finally, we can use SLAM-produced information about scene surfaces (e.g. standoff, orientation) and egomotion (e.g. yaw rate) to focus the system's text extraction efforts where they are likely to produce usable text rather than garbage. The technique enables text detection and decoding to run effectively at frame rate on the sensor's full surround, even though the CPU resources typically available on a mobile platform (robot, wearable or handheld device) are not sufficient to such methods on full images at sensor rates. Moreover, organizing detected text in a locally stable 3D frame enables combination of multiple noisy text observations into a single higher-confidence estimate of environmental text.
by Yafim Landa.
M. Eng.

The mapping problem is a major player in mobile robotics, and it is essential for many real applications such as disaster response or nuclear decommissioning. Generally, the robotic mapping is addressed under the umbrella of simultaneous localization and mapping (SLAM). Several probabilistic techniques were developed in the literature to approach the SLAM problem, and despite the good performance, their convergence proof is only limited to linear Gaussian models. This thesis proposes an interval SLAM (i-SLAM) algorithm as a new approach that addresses the robotic mapping problem in the context of interval methods. The noise of the robot sensor is assumed bounded, and without any prior knowledge of its distribution, we specify soft conditions that guarantee the convergence of robotic mapping for the case of nonlinear models with non-Gaussian noise. A new theory about compact sets is developed in the context of real analysis to conclude such conditions. Then, a case study is presented where the performance of i-SLAM is compared to the probabilistic counterparts in terms of accuracy and efficiency. Moreover, this work presents an application for i-SLAM using an RGB-D sensor that operates in unknown environments. Interval methods and computer vision techniques are employed to extract planar landmarks in the environment. Then, a new hybrid data association approach is developed using a modified version of bag-of-features method to uniquely identify different landmarks across timesteps. Finally, the results obtained using the proposed data association approach are compared to the typical least-squares approaches, thus demonstrating the consistency and accuracy of the proposed approach.

Ce mémoire décrit mes travaux de thèse de doctorat menés au sein de l’équipe ComSee (Computers that See) rattachée à l’axe ISPR (Image, Systèmes de Perception et Robotique) de l’Institut Pascal. Celle-ci a été financée par la Région Auvergne et le Fonds Européen de Développement Régional. Les travaux présentés s’inscrivent dans le cadre d’applications de localisation pour la robotique mobile et la Réalité Augmentée. Le framework réalisé au cours de cette thèse est une approche générique pour l’implémentation d’applications de SLAM : Simultaneous Localization And Mapping (algorithme de localisation par rapport à un modèle simultanément reconstruit). L’approche intègre une multitude de contraintes dans les processus de localisation et de reconstruction. Ces contraintes proviennent de données capteurs mais également d’a priori liés au contexte applicatif. Chaque contrainte est utilisée au sein d’un même algorithme d’optimisation afin d’améliorer l’estimation du mouvement ainsi que la précision du modèle reconstruit. Trois problèmes ont été abordés au cours de ce travail. Le premier concerne l’utilisation de contraintes sur le modèle reconstruit pour l’estimation précise d’objets 3D partiellement connus et présents dans l’environnement. La seconde problématique traite de la fusion de données multi-capteurs, donc hétérogènes et asynchrones, en utilisant un unique algorithme d’optimisation. La dernière problématique concerne la génération automatique et efficace d’algorithmes d’optimisation à contraintes multiples. L’objectif est de proposer une solution temps réel 1 aux problèmes de SLAM à contraintes multiples. Une approche générique est utilisée pour concevoir le framework afin de gérer une multitude de configurations liées aux différentes contraintes des problèmes de SLAM. Un intérêt tout particulier a été porté à la faible consommation de ressources (mémoire et CPU) tout en conservant une grande portabilité. De plus, la méta-programmation est utilisée pour générer automatiquement et spécifiquement les parties les plus complexes du code en fonction du problème à résoudre. La bibliothèque d’optimisation LMA qui a été développée au cours de cette thèse est mise à disposition de la communauté en open-source. Des expérimentations sont présentées à la fois sur des données de synthèse et des données réelles. Un comparatif exhaustif met en évidence les performances de la bibliothèque LMA face aux alternatives les plus utilisées de l’état de l’art. De plus, le framework de SLAM est utilisé sur des problèmes impliquant une difficulté et une quantité de contraintes croissantes. Les applications de robotique mobile et de Réalité Augmentée mettent en évidence des performances temps réel et un niveau de précision qui croît avec le nombre de contraintes utilisées
This report describes my thesis work conducted within the ComSee (Computers That See) team related to the ISPR axis (ImageS, Perception Systems and Robotics) of Institut Pascal. It was financed by the Auvergne Région and the European Fund of Regional Development. The thesis was motivated by localization issues related to Augmented Reality and autonomous navigation. The framework developed during this thesis is a generic approach to implement SLAM algorithms : Simultaneous Localization And Mapping. The proposed approach use multiple constraints in the localization and mapping processes. Those constraints come from sensors data and also from knowledge given by the application context. Each constraint is used into one optimization algorithm in order to improve the estimation of the motion and the accuracy of the map. Three problems have been tackled. The first deals with constraints on the map to accurately estimate the pose of 3D objects partially known in the environment. The second problem is about merging multiple heterogeneous and asynchronous data coming from different sensors using an optimization algorithm. The last problem is to write an efficient and real-time implementation of the SLAM problem using multiple constraints. A generic approach is used to design the framework and to generate different configurations, according to the constraints, of each SLAM problem. A particular interest has been put in the low computational requirement (in term of memory and CPU) while offering a high portability. Moreover, meta-programming techniques have been used to automatically and specifically generate the more complex parts of the code according to the given problem. The optimization library LMA, developed during this thesis, is made available of the community in open-source. Several experiments were done on synthesis and real data. An exhaustive benchmark shows the performances of the LMA library compared to the most used alternatives of the state of the art. Moreover, the SLAM framework is used on different problems with an increasing difficulty and amount of constraints. Augmented Reality and autonomous navigation applications show the good performances and accuracies in multiple constraints context

This thesis focuses on the problem of LiDAR sensor-based mapping where conventional methods has difficulties with the long-term operation or sensor integration. A new mapping system framework has been proposed to overcome the shortcomings of the conventional methods and we demonstrate its advantages on multiple map datasets collected from various environments. The outcome of the research will be useful for several applications where long-term mapping required, such as security robots, autonomous cars, and service robots.

In robotics the Simultaneous Localization and Mapping (SLAM) is the problem in which an autonomous robots acquires a map of the surrounding environment while at the same time localizes itself inside this map. In the last years a lot of researchers have spent a great effort in developing new families of algorithms, using several sensors and robotic platforms. One of the most challenging field of research in SLAM is the so called Visual-SLAM problem, in which various types of cameras are used as sensor for the navigation. Cameras are inexpensive sensors and can provide rich information about the surrounding environment, on the other hand the complexity of the computer vision tasks and the strong dependence on the characteristics of the environment in current approaches makes the Visual-SLAM far to be considered a closed problem. Most of the SLAM algorithm are usually tested on wheeled robot. These platforms have become robust and stable, on the other hand the research in robot design moves toward a new family of robot platforms, the humanoid robots. Just like humans, a humanoid robot can adapt itself to changes in the environment in order to efficiently reach its goals. Despite that, only a few roboticists focused theirs research on stable implementation of SLAM and Visual SLAM algorithms well suited for humanoid robots. Humanoid platforms raise issues which can compromise the stability of the conventional navigation algorithms, especially for vision-based approaches. A humanoid robot can move in 3D without the usual planar motion assumption that constraint the movement in 2D, usually with quick and complex movements combined with unpredictable vibrations, compromising the reliability of the acquired sensors data, for example introducing in the images grabbed by the camera an undesired motion blur effect. Due to the strong balance constraints, a humanoid robot usually can’t be equipped with powerfull but hefty computer boards: this limits the implementation of complex and computational expensive algorithms. Moreover, unlike wheeled robots, its complex kinematics usually forbids a reliable reconstruction of the motion from the servo-motor encoders. In this thesis, we focus on studying and developing new techniques addressing the Visual-SLAM problem, with particular attention to the issues related to using as experimental platform small humanoid robots equipped with a single perspective camera. The main efforts in SLAM and Visual SLAM research areas have been put into the estimation functionality. However, most of the functionalities involved in Visual SLAM are in perception processes. In this thesis we therefore focus on the improvement of the perceptual processes, from a computer vision point-of-view. We faced small humanoid robot related issues like low-computational capability, the low quality of the sensor data and the high degrees of freedom of the motion. We cope with the low computational resources presenting a new similarity measure for images based on a compact signature to be used in image-based topological SLAM problem. The motion blur problem is faced proposing a new feature detection and tracking scheme that is robust even to non-uniform motion blur. We develop a framework for visual odometry based on features robust to motion blur. We finally propose an homography-based approach to 3D visual SLAM, using the information provided by a single camera mounted on a humanoid robot, based on the assumption that the robot moves on a planar environment. All proposed methods have been validated with experiments and comparative validation using both standard datasets and images taken by the cameras mounted on walking small humanoid robots.
Nell’ambito della robotica, il Simultaneous Localization and Mapping (SLAM) é il processo grazie al quale un robot autonomo é in grado di creare una mappa dell’ambiente circostante e allo stesso tempo di localizzarsi avvalendosi di tale mappa. Negli ultimi anni un considerevole numero di ricercatori ha sviluppato nuove famiglie di algoritmi di SLAM, basati su vari sensori e utilizzando varie piattaforme robotiche. Uno degli ambiti più complessi nella ricerca sullo SLAM é il cosiddetto Visual-SLAM, che prevede l’utilizzo di vari tipi di telecamera come sensore per la navigazione. Le telecamere sono sensori economici che raccolgono molte informazioni sull’ambiente circostante. D’altro canto, la complessità degli algoritmi di visione artificiale e la forte dipendenza degli approcci attualmente realizzati dalle caratteristiche dell’ambiente, rendono il Visual-SLAM un problema lontano dal poter essere considerato risolto. Molti degli algoritmi di SLAM sono solitamente testati usando robot dotati di ruote. Sebbene tali piattaforme siano ormai robuste e stabili, la ricerca sulla progettazione di nuove piattaforme robotiche sta in parte migrando verso la robotica umanoide. Proprio come gli esseri umani, i robot umanoidi sono in grado di adattarsi ai cambiamenti dell’ambiente per raggiungere efficacemente i propri obiettivi. Nonostante ciò, solo pochi ricercatori hanno focalizzato i loro sforzi su implementazioni stabili di algoritmi di SLAM e Visual-SLAM adatti ai robot umanoidi. Tali piattaforme robotiche introducono nuove problematiche che possono compromettere la stabilità degli algoritmi di navigazione convenzionali, specie se basati sulla visione. I robot umanoidi sono dotati di un alto grado di libertà di movimento, con la possibilità di effettuare velocemente movimenti complessi: tali caratteristiche introducono negli spostamenti vibrazioni non deterministiche in grado di compromettere l’affidabilit` dei dati sensoriali acquisiti, per esempio introducendo nei flussi video effetti indesiderati quali il motion blur. A causa dei vincoli imposti dal bilanciamento del corpo, inoltre, tali robot non sempre possono essere dotati di unit` di elaborazione molto performanti che spesso sono ingombranti e dal peso elevato: ci` limita l’utilizzo di algoritmi complessi e computazionalmente gravosi. Infine, al contrario di quanto accade per i robot dotati di ruote, la complessa cinematica di un robot umanoide impedisce di ricostruire il movimento basandosi sulle informazioni provenienti dagli encoder posti sui motori. In questa tesi ci si é focalizzati sullo studio e sullo sviluppo di nuove metodologie per affrontare il problema del Visual-SLAM, ponendo particolare enfasi ai problemi legati all’utilizzo di piccoli robot umanoidi dotati di una singola telecamera come piattaforme per gli esperimenti. I maggiori sforzi nell’ambito della ricerca sullo SLAM e sul Visual-SLAM si sono concentrati nel campo del processo di stima dello stato del robot, ad esempio la stima della propria posizione e della mappa dell’ambiente. D’altra parte, la maggior parte delle problematiche incontrate nella ricerca sul Visual-SLAM sono legate al processo di percezione, ovvero all’interpretazione dei dati provenienti dai sensori. In questa tesi ci si é perciò concentrati sul miglioramento dei processi percettivi da un punto di vista della visione artificiale. Sono stati affrontati i problemi che scaturiscono dall’utilizzo di piccoli robot umanoidi come piattaforme sperimentali, come ad esempio la bassa capacità di calcolo, la bassa qualit` dei dati sensoriali e l’elevato numero di gradi di libertà nei movimenti. La bassa capacità di calcolo ha portato alla creazione di un nuovo metodo per misurare la similarità tra le immagini, che fa uso di una descrizione dell’immagine compatta, utilizzabile in applicazioni di SLAM topologico. Il problema del motion blur é stato affrontato proponendo una nuova tecnica di rilevamento di feature visive, unitamente ad un nuovo schema di tracking, robusto an- che in caso di motion blur non uniforme. E’ stato altresì sviluppato un framework per l’odometria basata sulle immagini, che fa uso delle feature visive presentate. Si propone infine un approccio al Visual-SLAM basato sulle omografie, che sfrutta le informazioni ottenute da una singola telecamera montata su un robot umanoide. Tale approccio si basa sull’assunzione che il robot si muove su una superficie piana. Tutti i metodi proposti sono stati validati con esperimenti e studi comparativi, usando sia dataset standard che immagini acquisite dalle telecamere installate su piccoli robot umanoidi.

Cette thèse traite du problème de localisation et contrôle de robots humanoïdes par rapport à leur environnement, tel qu'observé par ses capteurs embarqués. Le SLAM visuel dense, consistant en l'estimation simultanée d'une carte 3D de l'environnement et de la position du robot dans cette carte est exploité pour étendre et robustifier les méthodes de planification contrôle multi-contact. Celles-ci consistent à établir et exploiter des contacts robot-environnement pour accomplir des tâches de locomotion et manipulation. Des incertitudes sur la posture initiale du robot, ainsi que des perturbations causées par une modélisation inadéquate des contacts, ainsi que des perturbations externes oblige à la prise en compte de l'état du robot et son environnement. Une méthode de calibration corps-complet est également proposée, afin d'obtenir une connaissance fiable de la chaîne cinématique du robot, nécessaire pour réaliser de telles tâches. Finalement, une méthode de marche basée sur de la commande prédictive de modèles est robustifiée par la prise en compte de large perturbations, permettant d'ajuster les trajectoires de pied et du centre de masse afin de garantir sa stabilité, tout en accomplissant les objectifs désirés. Les méthodes proposées sont illustrées et validées par de multiples expérimentations sur les robots humanoïdes HRP-2Kai et HRP-4
This thesis deals with the problem of localizing and controlling humanoid robots with respect to its environment, as observed by its on-board sensors. Dense visual SLAM, consisting in the simultaneous estimation of a 3D map of the environment and of the robot localization within that maps is exploited to extend and robustify multi-contact planning and control. Establishing and exploiting robot-environment contacts allows the accomplishment of both locomotion and manipulation tasks. Uncertainties in the initial robot posture, and perturbations arising from improper contact-modelling and external causes are accounted for by observing the state of the robot and its environment. A whole-body calibration method is also proposed, so that robust knowledge of the robot's kinematic structure is known, a prerequisite to all robot-environment interaction tasks. Finally, a walking method based on model predictive control is robustified by taking into account large perturbations, and adjusting the footstep and center-of-mass trajectories accordingly to guarantee stability while accomplishing desired objectives.Several experiments on an HRP-2Kai and an HRP-4 humanoid robots are presented and discussed to illustrate and validate each of the proposed methods

Examensarbetet är utfört i Rottneros bruks barkpanna och syftet med arbetet är att genom provkörningar och beräkningar, tillslut finna hur mycket slam som är praktiskt möjligt att blanda in i bränslemixen till pannan. Intresset i att elda slam är idag väldigt stort hos producenter av slam, då istället för att skicka iväg slammet för hantering kan en förbränning hos dem själva istället göras. Det stora intresset i att elda slammet själv, ligger i att på det sättet kan stora ekonomiska besparingar göras. Eftersom Rottneros bruk får slam som restprodukt från deras reningsverk, blir detta även aktuellt hos dem. Rottneros Bruk har idag ett företag som kommer och hämtar allt slam, behovet för detta skulle kunna efter denna studie minska.   Målet med studien är att hitta en bränslemix till pannan med så stor slamhalt som är praktiskt möjlig. Detta för att den ekonomiska besparingen ska bli så stor som möjligt, då besparingen ökar med andelen slam som går att blanda in i bränslemixen. För att en maximalt fungerande slammängd i bränslemixen ska vara acceptabel, krävs det att både bränslemixen är körvänlig i pannan samt att kraven på utsläppen uppfylls. Körbarheten innefattar att effekten, temperaturen samt slaggbildningen i pannan ligger på en sådan nivå att de inte blir problematiskt. När det kommer till miljöaspekten krävs det att så kallade gränsvärden för NOx och CO utsläpp inte överskrids. Dessa gränsvärden är unika för Rottneros bruk, som i sin tur får konsekvenser i form av böter ifall dem överskrids.        För att kunna lösa detta har provkörningar och beräkningar gjorts hos pannan.  Provkörningarna som genomförts har haft olika slammängder samt torrhalter på bränslemixen, för att på detta sätt se hur pannan reagerar under olika förhållanden. Genom beräkningar och data ifrån provkörningarna kan ett framtagande göras för hur mycket slam som går att blanda in i bränslemixen.   Det resultat som fås från denna studie säger att en fungerande bränslemix till pannan kan ha en maximal slaminblandning på 1,5-2 procent. Med denna mängd slam i bränslemixen kommer en årlig besparing för Rottneros bruk ligga på 207,6-276,9 kkr/år. Bränslemixen kommer både uppfylla gränsvärdena för utsläpp av NOx och CO samt att körbarheten i pannan kommer vara acceptabel. Den låga inbladningen av slam beror först och främst på att utsläppen av NOx annars kommer överskridas, detta på grund av att kvävet i slammet. Torrhalten hos bränslet har även ett stort inflytande på hur stora utsläppen blir av NOx och CO. Då torrhalten hos slammet, flisen och barken varierar, är detta något som hänsyn har tagits till då den maximala slaminblandingen hos bränslemixen tagits fram.

Lo scopo di questa tesi è quello di introdurre la tecnologia radar ad onde millimetriche applicata a problemi di mapping e di localizzazione in modo da investigare la fattibilità del recente concetto di "personal radar". Più specificatamente, può essere inteso come una caratteristica futura degli apparecchi mobili per la stima delle mappe degli ambienti interni sfruttando le tecnologie di comunicazione già presenti, avendo così la possibilità di creare applicazioni di localizzazione che non necessitano di infrastrutture ad-hoc. Il personal radar scansiona automaticamente l'ambiente circostante e mediante l'utilizzo di antenne colleziona le risposte provenienti dall'ambiente ad ogni direzione di scansione. Successivamente analizzando i dati collezionati il personal radar è in grado di dedurre la mappa dell'ambiente. Al momento le tecnologie più usate sono quella lidar o quella camera-based ma sono solitamente tecnologie più costose e che richiedono supporti meccanici e perfette condizioni di visibilità dell'ambiente. Per queste ragioni è interessante esplorare la tecnologia radar ad onde millimetriche. Questa teconologia infatti offre la possiilità di impacchettare un numero elevato di antenne in un piccolo spazio e quindi di realizzare diagrammi di radiazione molto stretti alle spese di una degradazione delle perfomance. Per queste ragioni è stato deciso di suddividere la tesi nei seguenti capitoli. Nel primo capitolo viene fornita una breve spiegazione della teoria radar con attenzione particolare alla tecnologia (FMCW) del dispositivo utilizzato. Nel secondo capitolo viene fornita una analisi della teoria dello SLAM con un approfondimento sulla teoria SLAM basata su grafi. Dopodichè è presente una analisi del dispositivo utilizzato. Nel quarto capitolo vengono descritti gli algoritmi sviluppati e i motivi per cui sono stati necessari e infine nel capitolo finale vengono riportati i risultati finali con le relative considerazioni.

Navigation with unmanned aerial vehicles (UAVs) requires good knowledge of the current position and other states. A UAV navigation system often uses GPS and inertial sensors in a state estimation solution. If the GPS signal is lost or corrupted state estimation must still be possible and this is where simultaneous localization and mapping (SLAM) provides a solution. SLAM considers the problem of incrementally building a consistent map of a previously unknown environment and simultaneously localize itself within this map, thus a solution does not require position from the GPS receiver.

This thesis presents a visual feature based SLAM solution using a low resolution video camera, a low-cost inertial measurement unit (IMU) and a barometric pressure sensor. State estimation in made with a extended information filter (EIF) where sparseness in the information matrix is enforced with an approximation.

An implementation is evaluated on real flight data and compared to a EKF-SLAM solution. Results show that both solutions provide similar estimates but the EIF is over-confident. The sparse structure is exploited, possibly not fully, making the solution nearly linear in time and storage requirements are linear in the number of features which enables evaluation for a longer period of time.

Sweden's largest sewage treatment plant, Henrikdal, will be rebuilt to receive twice as much sludge than they can do today. To have the ability to do that, either the level of total solids (TS) must increase or expand the rotary chambers. Off which the second option is to be avoided due to the high costs.The report will answer why the origin, viscosity, shear rate, temperature and TS-level of the sludge is important to consider when pumping it. Also, the report will show results of how the viscosity will get higher either when the TS-level or the shear rate is increasing as well as the temperature is decreasing. Beyond that, the report will contain source gathering, method, conclusion, discussion and ends with proposals for further research.

Nos últimos anos, observamos o alvorecer de uma grande quantidade de aplicações que utilizam robôs autônomos. Para que um robô seja considerado verdadeiramente autônomo, é primordial que ele possua a capacidade de aprender sobre o ambiente no qual opera. Métodos de SLAM (Localização e Mapeamento Simultâneos) constroem um mapa do ambiente por onde o robô trafega ao mesmo tempo em que estimam a trajetória correta do robô. No entanto, para obter um mapa completo do ambiente de forma autônoma é preciso guiar o robô por todo o ambiente, o que é feito no problema de exploração. Câmeras são sensores baratos que podem ser utilizadas para a construção de mapas 3D. Porém, o problema de exploração em mapas gerados por métodos de SLAM monocular, i.e. que extraem informações de uma única câmera, ainda é um problema em aberto, pois tais métodos geram mapas esparsos ou semi-densos, que são inadequados para navegação e exploração. Para tal situação, é necessário desenvolver métodos de exploração capazes de lidar com a limitação das câmeras e com a falta de informação nos mapas gerados por SLAMs monoculares. Propõe-se uma estratégia de exploração que utilize mapas volumétricos locais, gerados através das linhas de visão, permitindo que o robô navegue em segurança. Nestes mapas locais, são definidos objetivos que levem o robô a explorar o ambiente desviando de obstáculos. A abordagem proposta visa responder a questão fundamental em exploração: "Para onde ir?". Além disso, busca determinar corretamente quando o ambiente está suficientemente explorado e a exploração deve parar. A abordagem proposta é avaliada através de experimentos em um ambiente simples (i.e. apenas uma sala) e em um ambiente compostos por diversas salas.
In recent years, we have seen the dawn of a large number of applications that use autonomous robots. For a robot to be considered truly autonomous, it is primordial that it has the ability to learn about the environment in which it operates. SLAM (Simultaneous Location and Mapping) methods build a map of the environment while estimating the robot’s correct trajectory. However, to autonomously obtain a complete map of the environment, it is necessary to guide the robot throughout the environment, which is done in the exploration problem. Cameras are inexpensive sensors that can be used for building 3D maps. However, the exploration problem in maps generated by monocular SLAM methods (i.e. that extract information from a single camera) is still an open problem, since such methods generate sparse or semi-dense maps that are ill-suitable for navigation and exploration. For such a situation, it is necessary to develop exploration methods capable of dealing with the limitation of the cameras and the lack of information in the maps generated by monocular SLAMs. We proposes an exploration strategy that uses local volumetric maps, generated using the lines of sight, allowing the robot to safely navigate. In these local maps, objectives are defined to lead the robot to explore the environment while avoiding obstacles. The proposed approach aims to answer the fundamental question in exploration: "Where to go?". In addition, it seeks to determine correctly when the environment is sufficiently explored and the exploration must stop. The effectiveness of the proposed approach is evaluated in experiments on single and multi-room environments.

The thesis tackles the problem of data association for monocular object-based SLAM, which gets often omitted in related works. A method for estimatingellipsoid object landmark representations is implemented. This method uses bounding box multi-view object detections from 2D images with the help ofYOLOv3 object detector and ORB-SLAM2 for camera pose estimation. The online data association uses SIFT image feature matching and landmark backprojectionmatching against bounding box detections to associate these object detections. This combination and its evaluation is the main contribution of the thesis. The overall algorithm is tested on several datasets, both real-world and computer rendered. The association algorithm manages well on the testedsequences and it is shown that matching with the back projections of the ellipsoidlandmarks improves the robustness of the approach. It is shown that with some implementation changes, the algorithm can run at real-time. The landmarkestimation part works satisfactory for landmark initialization. Based on the findings future work is proposed.
Examensarbetet tar upp problemet med dataassociation för monokulärt objektbaserat SLAM, som ofta utelämnas i relaterade verk. En metod för att uppskatta ellipsoida landmärkesrepresentationer implementeras. Den här metoden använder avgränsningsrutor med flera vyer i 2D-bilder med hjälp av YOLOv3-objektdetektor och ORB-SLAM2 för uppskattning av kamerapositioner. Onlinedataassociationen använder SIFT intressepunktsmatchning samt matchnin gav landmärkesbakprojektioner mot avgränsningsrutorna för att associera dessa objektdetektioner. Denna kombination och dess utvärdering är huvudbidraget i examensarbetet. Den övergripande algoritmen testas på flera datasätt, bådeverkliga och datorgenererade. Associeringsalgoritmen hanterar väl de testa desekvenserna och det visas att matchning med bakprojektionerna av ellipsoidalandmärken gör associeringen mer robust. Det visas att med vissa implementeringsförändringarkan algoritmen köras i realtid. Landmarkeringsberäkningsdelen fungerar acceptabelt för initiering av landmärken. Baserat på resultaten föreslås framtid arbete.

The thesis tackles the problem of data association for monocular object-basedSLAM, which gets often omitted in related works. A method for estimating ellipsoid object landmark representations is implemented. This method uses bounding box multi-view object detections from 2D images with the help ofYOLOv3 object detector and ORB-SLAM2 for camera pose estimation. The online data association uses SIFT image feature matching and landmark back projection matching against bounding box detections to associate these object detections. This combination and its evaluation is the main contribution of the thesis. The overall algorithm is tested on several datasets, both real-world and computer rendered. The association algorithm manages well on the tested sequences and it is shown that matching with the back projections of the ellipsoid landmarks improves the robustness of the approach. It is shown that with some implementation changes, the algorithm can run at real-time. The landmark estimation part works satisfactory for landmark initialization. Based on the findings future work is proposed.
Examensarbetet tar upp problemet med dataassociation för monokulärt objektbaserat SLAM, som ofta utelämnas i relaterade verk. En metod för att uppskatta ellipsoida landmärkesrepresentationer implementeras. Den här metodenanvänder avgränsningsrutor med flera vyer i 2D-bilder med hjälp av YOLOv3-objektdetektor och ORB-SLAM2 för uppskattning av kamerapositioner. Online dataassociationen använder SIFT intressepunktsmatchning samt matchningav landmärkesbakprojektioner mot avgränsningsrutorna för att associera dessaobjektdetektioner. Denna kombination och dess utvärdering är huvudbidrageti examensarbetet. Den övergripande algoritmen testas på flera datasätt, bådeverkliga och datorgenererade. Associeringsalgoritmen hanterar väl de testadesekvenserna och det visas att matchning med bakprojektionerna av ellipsoidalandmärken gör associeringen mer robust. Det visas att med vissa implementeringsförändringar kan algoritmen köras i realtid. Landmarkeringsberäkningsdelen fungerar acceptabelt för initiering av landmärken. Baserat på resultatenföreslås framtid arbete.

The problem of autonomous navigation of a mobile device is at the heart of the more general issue of spatial awareness and is now a well-studied problem in the robotics community. Following a plethora of approaches throughout the history of this research, recently, implementations have been converging towards vision-based methods. While the primary reason for this success is the enormous amount of information content encrypted in images, this is also the main obstacle in achieving faster and better solutions. The growing demand for high-performance systems able to run on affordable hardware pushes algorithms to the limits, imposing the need for more effective approximations within the estimation process. The biggest challenge lies in achieving a balance between two competing goals: the optimisation of time complexity and the preservation of the desired precision levels. The key is in agile manipulation of data, which is the main idea explored in this thesis. Exploiting the power of probabilistic priors in sequential tracking, we conduct a theoretical investigation of the information encoded in measurements and estimates, which provides a deep understanding of the map structure as perceived through the camera lens. Employing information theoretic principles to guide the decisions made throughout the estimation process we demonstrate how this methodology can boost both the efficiency and consistency of algorithms. Focusing on the most challenging processes in a state of the art system, we apply our information theoretic framework to local motion estimation and maintenance of large probabilistic maps. Our investigation gives rise to dynamic algorithms for quality map-partitioning and robust feature mapping in the presence of significant ambiguity and variable camera dynamics. The latter is further explored to achieve scalable performance allowing dense feature matching based on concrete probabilistic decisions.

Magíster en Ciencias de la Ingeniería, Mención Eléctrica. Ingeniero Civil Eléctrico
Este trabajo trata sobre el diseño e implementación de un sistema de Localización y Mapeo Simultáneo (SLAM) visual usando la teoría de Conjuntos Finitos Aleatorios (RFS), en el que un navegador (e.g. robot, auto, teléfono celular, etc.) utiliza una cámara de vídeo RGB-D para reconstruir la escena a su alrededor y al mismo tiempo descubrir su propia posición. Esta capacidad es relevante para las tecnologías del futuro, que deberán desplazarse sin ayuda externa. Se considera la inclusión de modelos realistas de medición y movimiento, incluyendo la intermitencia de las detecciones de objetos, la presencia de falsos positivos en las mediciones y el ruido en la imagen. Para ello se analizan sistemas basados en la teoría RFS, que es capaz de incluir estos efectos de manera fundamentada, a diferencia de otras alternativas del estado del arte que se basan en heurísticas poco realistas como el conocimiento absoluto de las asociaciones de datos entre mediciones y puntos en el mapa. Se incluye una amplia revisión de la literatura, desde Structure from Motion a Odometría Visual, a los distintos algoritmos para SLAM. Luego, se procede a explicar los detalles de implementación de un sistema flexible para el análisis de algoritmos de SLAM, así como la implementación particular del algoritmo Rao-Blackwellized (RB)-Probability Hypothesis Density (PHD)-SLAM. Se presentan análisis del desempeño de este algoritmo al cambiar las distintas estadísticas que pueden variar en su uso práctico. Se hace una comparación detallada con la alternativa Incremental Smoothing and Mapping (iSAM2), usualmente usada en otros sistemas del estado del arte. Luego, basado en la teoría de Modelos Gráficos Probabilísticos (PGM) que está detrás de iSAM2, se propone un nuevo algoritmo, Loopy PHD-SLAM, capaz de propagar información a lo largo del grafo inducido de manera eficiente, incluyendo las estadísticas de RFS. Con una implementación sencilla como prueba de concepto, se observa la capacidad de este nuevo método de cerrar ciclos y converger a soluciones correctas.
Este trabajo ha sido auspiciado por Conicyt

Autonomous vehicles have grown into a hot topic in both research and industry. For a vehicle to be able to run autonomously, it needs several different types of systems to function properly. One of the most important of them is simultaneous localization and mapping (SLAM). It is used for estimating the pose of the vehicle and building a map of the environment around it based on sensor readings. In this thesis we have developed an novel approach to measure and evaluate the quality of a landmark-based SLAM algorithm in a static environment. The error measurement evaluation is a multi-error function and consists of the following error types: average pose error, maximum pose error, number of false negatives, number of false positives and an error relating to the distance when landmarks are added into the map. The error function can be tailored towards specific applications by settings different weights for each error. A small research concept car with several different sensors and an outside tracking system was used to create several datasets. The datasets include three different map layouts and three different power settings on the car’s engine to create a large variability in the datasets. FastSLAM and EKF-SLAM were to test the proposed SLAM evaluation method. A comparison to just the pose error was made to asses if our method can provide more information concerning establishing SLAM quality. Our results show that the pose error is often a good enough indicator of SLAM quality. However it can occasionally be misleading with errors related to mapping (location of landmarks, false negative and false positive landmarks). By using the method presented in this thesis, errors relating to the mapping will be more easily detected than by looking at the pose error.
Autonoma fordon har vuxit till ett viktigt ämne både inom forskning och industri. För att ett fordon ska kunna köras autonomt behövs det att flera olika system fungerar korrekt. En av de viktigaste av dem är simultaneous localization and mapping (SLAM). Det används för att uppskatta fordonets position samt för att bygga en karta av miljön runt fordonet. I det här examensarbetet har vi utvecklat en ny metod för att mäta och utvärdera kvaliteten på en landmärkes-baserad SLAM-algoritm i en statisk miljö. Metoden består av följande feltyper: medelvärdet av positionsfelet, det maximala positionsfelet, antal falskt negativa fel, antal falskt positiva fel samt ett fel relaterat till avståndet när ett landmärke läggs till i kartan. Genom att använda vikter för varje fel kan metoden skräddarsys till en specifik applikation. En liten konceptbil med flera olika sensorer och ett yttre spårningssystem användes för att skapa flera dataset. Dataseten innehåller tre olika kartlayouter och tre olika effektinställningar på bilen för att skapa stor variation i dataseten. FastSLAM och EKF-SLAM användes vid testningen av den nya metoden för kvalitetsbedömning av SLAM. Den nya metoden jämfördes mot positionsfelet för att analysera ifall den nya metoden är ett bättre sätt att mäta SLAMkvalitet. Våra resultat visar att positionsfelet ofta är en tillräckligt bra indikator för SLAM-kvalitet, men det kan ibland vara vilseledande gällande fel i kartläggningen (positioner av landmärken, falskt negativa fel och falskt positiva fel). Genom att använda metoden som presenteras i det här examensarbete är fel som är relaterade till kartläggningen lättare att upptäcka än om man enbart kollar på positionsfelet.

Cette thèse propose des modèles cartographiques à grande échelle d'environnements urbains et ruraux à l'aide de données en 3D acquises par plusieurs robots. La mémoire contribue de deux manières principales au domaine de recherche de la cartographie. La première contribution est la création d'une nouvelle structure, CoMapping, qui permet de générer des cartes 3D de façon collaborative. Cette structure s’applique aux environnements extérieurs en ayant une approche décentralisée. La fonctionnalité de CoMapping comprend les éléments suivants : Tout d’abord, chaque robot réalise la construction d'une carte de son environnement sous forme de nuage de points.Pour cela, le système de cartographie a été mis en place sur des ordinateurs dédiés à chaque voiture, en traitant les mesures de distance à partir d'un LiDAR 3D se déplaçant en six degrés de liberté (6-DOF). Ensuite, les robots partagent leurs cartes locales et fusionnent individuellement les nuages de points afin d'améliorer leur estimation de leur cartographie locale. La deuxième contribution clé est le groupe de métriques qui permettent d'analyser les processus de fusion et de partage de cartes entre les robots. Nous présentons des résultats expérimentaux en vue de valider la structure CoMapping et ses métriques. Tous les tests ont été réalisés dans des environnements extérieurs urbains du campus de l’École Centrale de Nantes ainsi que dans des milieux ruraux
This thesis proposes large-scale mapping model of urban and rural environments using 3D data acquired by several robots. The work contributes in two main ways to the research field of mapping. The first contribution is the creation of a new framework, CoMapping, which allows to generate 3D maps in a cooperative way. This framework applies to outdoor environments with a decentralized approach. The CoMapping's functionality includes the following elements: First of all, each robot builds a map of its environment in point cloud format.To do this, the mapping system was set up on computers dedicated to each vehicle, processing distance measurements from a 3D LiDAR moving in six degrees of freedom (6-DOF). Then, the robots share their local maps and merge the point clouds individually to improve their local map estimation. The second key contribution is the group of metrics that allow to analyze the merging and card sharing processes between robots. We present experimental results to validate the CoMapping framework with their respective metrics. All tests were carried out in urban outdoor environments on the surrounding campus of the École Centrale de Nantes as well as in rural areas

State-of-the-art LIDAR odometry techniques are exceptionally precise. However, while they solve the localization problem, they perform mapping on-the-run, not being able to close loops, neither re-localize in previously visited environments. This study is concerned with the development of a system that combines an accurate laser odometry estimator, with algorithms for place recognition in order to detect trajectory loops. This project uses widely available datasets from urban driving scenarios and outdoor areas for development and evaluation of the system The results obtained confirm that loop closure detection can significantly improve the accuracy and robustness of laser SLAM pipelines, with detectors based on point cloud segments and visual features displaying very strong performance during the evaluation phase.
Spjutspetsen inom Lidar-baserade teknik för fordonsodometri har den senaste tiden uppnått exceptionella nivåer av noggrannhet. Med det sagt har de metoder som presenterats fokuserat på att lösa lokaliseringsproblemet och därför gjort förenklande antaganden såsom att de sköter kartläggning av miljön löpande utan platsåterkoppling, och att de inte kan återlokalisera i tidigare kända miljöer. Således utvecklar vi i detta arbete ett system som kombinerar dessa noggranna lidarodometriska tekniker med algoritmer för platsigenkänning för att möjliggöra loopdetektion. Vi använder vitt tillgängliga dataset av körning i stadstrafik samt i utomhusområden för utveckling och utvärdering av systemet. Resultaten visar att platsåterkoppling förbättrar noggrannheten hos Lidar-baserade lokaliseringsmetoder och gör dem mer robusta, samt att man med hjälp av detektorer baserade på punktmolnssegmentering och visuella särdrag erhåller ett system som uppvisar mycket goda resultat under utvärderingsfasen.

La navigation autonome est un axe de recherche principal dans le domaine de la robotique mobile. Dans ce contexte, le robot doit disposer des algorithmes qui lui permettent d’évoluer de manière autonome dans des environnements complexes et inconnus. Les algorithmes de SLAM permettent à un robot de cartographier son environnement tout en se localisant dans l’espace. Les algorithmes SLAM sont de plus en plus performants, mais aucune implémentation matérielle ou architecturale complète n’a eu. Une telle implantation d’architecture doit prendre en considération la consommation d’énergie, l’embarquabilité et la puissance de calcul. Ce travail scientifique vise à évaluer des systèmes embarqués impliquant de la localisation ou reconstruction de scène. La méthodologie adoptera une approche A3 (Adéquation Algorithme Architecture) pour améliorer l’efficacité de l’implantation des algorithmes plus particulièrement pour des systèmes à fortes contraintes. Le système SLAM embarqué doit disposer d’une architecture électronique et logicielle permettant d’assurer la production d’information pertinentes à partir de données capteurs, tout en assurant la localisation de l’embarquant dans son environnement. L’objectif est donc de définir, pour un algorithme choisi, un modèle d’architecture répondant aux contraintes de l’embarqué. Les premiers travaux de cette thèse ont consisté à explorer les différentes approches algorithmiques permettant la résolution du problème de SLAM. Une étude plus approfondie de ces algorithmes est réalisée. Ceci nous a permet d’évaluer quatre algorithmes de différente nature : FastSLAM2.0, ORB SLAM, RatSLAM et le SLAM linéaire. Ces algorithmes ont été ensuite évalués sur plusieurs architectures pour l’embarqué afin d’étudier leur portabilité sur des systèmes de faible consommation énergétique et de ressources limitées. La comparaison prend en compte les temps d’exécutions et la consistance des résultats. Après avoir analysé profondément les évaluations temporelles de chaque algorithme, le FastSLAM2.0 est finalement choisi, pour un compromis temps d’exécution-consistance de résultat de localisation, comme candidat pour une étude plus approfondie sur une architecture hétérogène embarquée. La second partie de cette thèse est consacré à l’étude d’un système embarqué implémentant le FastSLAM2.0 monoculaire dédié aux environnements larges. Une réécriture algorithmique du FastSLAM2.0 a été nécessaire afin de l’adapter au mieux aux contraintes imposées par les environnements de grande échelle. Dans une démarche A3, le FastSLAM2.0 a été implanté sur une architecture hétérogène CPU-GPU. Grâce à un partitionnement efficace, un facteur d’accélération global de l’ordre de 22 a été obtenu sur une architecture récente dédiée pour l’embarqué. La nature du traitement de l’algorithme FastSLAM2.0 pouvait bénéficier d’une architecture fortement parallèle. Une deuxième instance matérielle basée sur une architecture programmable FPGA est proposée. L’implantation a été réalisée en utilisant des outils de synthèse de haut-niveau afin de réduire le temps de développement. Une comparaison des résultats d’implantation sur cette architecture matérielle par rapport à des architectures à base de GPU a été réalisée. Les gains obtenus sont conséquent, même par rapport aux GPU haut-de-gamme avec un grand nombre de cœurs. Le système résultant peut cartographier des environnements larges tout en garantissant le compromis entre la consistance des résultats de localisation et le temps réel. L’utilisation de plusieurs calculateurs implique d’utiliser des moyens d’échanges de données entre ces derniers. Cela passe par des couplages forts. Ces travaux de thèse ont permis de mettre en avant l’intérêt des architectures hétérogènes parallèles pour le portage des algorithmes SLAM. Les architectures hétérogènes à base de FPGA peuvent particulièrement devenir des candidats potentiels pour porter des algorithmes complexes traitant des données massives
Autonomous navigation is a main axis of research in the field of mobile robotics. In this context, the robot must have an algorithm that allow the robot to move autonomously in a complex and unfamiliar environments. Mapping in advance by a human operator is a tedious and time consuming task. On the other hand, it is not always reliable, especially when the structure of the environment changes. SLAM algorithms allow a robot to map its environment while localizing it in the space.SLAM algorithms are becoming more efficient, but there is no full hardware or architectural implementation that has taken place . Such implantation of architecture must take into account the energy consumption, the embeddability and computing power. This scientific work aims to evaluate the embedded systems implementing locatization and scene reconstruction (SLAM). The methodology will adopt an approach AAM ( Algorithm Architecture Matching) to improve the efficiency of the implementation of algorithms especially for systems with high constaints. SLAM embedded system must have an electronic and software architecture to ensure the production of relevant data from sensor information, while ensuring the localization of the robot in its environment. Therefore, the objective is to define, for a chosen algorithm, an architecture model that meets the constraints of embedded systems. The first work of this thesis was to explore the different algorithmic approaches for solving the SLAM problem. Further study of these algorithms is performed. This allows us to evaluate four different kinds of algorithms: FastSLAM2.0, ORB SLAM, SLAM RatSLAM and linear. These algorithms were then evaluated on multiple architectures for embedded systems to study their portability on energy low consumption systems and limited resources. The comparison takes into account the time of execution and consistency of results. After having deeply analyzed the temporal evaluations for each algorithm, the FastSLAM2.0 was finally chosen for its compromise performance-consistency of localization result and execution time, as a candidate for further study on an embedded heterogeneous architecture. The second part of this thesis is devoted to the study of an embedded implementing of the monocular FastSLAM2.0 which is dedicated to large scale environments. An algorithmic modification of the FastSLAM2.0 was necessary in order to better adapt it to the constraints imposed by the largescale environments. The resulting system is designed around a parallel multi-core architecture. Using an algorithm architecture matching approach, the FastSLAM2.0 was implemeted on a heterogeneous CPU-GPU architecture. Uisng an effective algorithme partitioning, an overall acceleration factor o about 22 was obtained on a recent dedicated architecture for embedded systems. The nature of the execution of FastSLAM2.0 algorithm could benefit from a highly parallel architecture. A second instance hardware based on programmable FPGA architecture is proposed. The implantation was performed using high-level synthesis tools to reduce development time. A comparison of the results of implementation on the hardware architecture compared to GPU-based architectures was realized. The gains obtained are promising, even compared to a high-end GPU that currently have a large number of cores. The resulting system can map a large environments while maintainingthe balance between the consistency of the localization results and real time performance. Using multiple calculators involves the use of a means of data exchange between them. This requires strong coupling (communication bus and shared memory). This thesis work has put forward the interests of parallel heterogeneous architectures (multicore, GPU) for embedding the SLAM algorithms. The FPGA-based heterogeneous architectures can particularly become potential candidatesto bring complex algorithms dealing with massive data

A commercially available camera, consisting of two separate 195 degree field of view fisheye lens cameras, is suggested for use in monocular SLAM. OpenCV’s fisheye camera model was corrected for viewing angles greater than 90 degrees and then implemented in ORB-SLAM2. Then, using the fisheye model as well as MultiCol-SLAM, we demonstrate the benefit of using wide-angle cameras for SLAM. The tracking provided by a wide-angle fisheye lens during sharp rotations outperforms that of conventional perspective cameras. We argue that so long as the camera model is sufficiently accurate, SLAM is able to perform well. By providing a complete image of the environment, Monocular SLAM will be robust to rotation.
En kommersiellt tillgänglig kamera, bestående av två separata kameror med fisheyelinser med 195 graders synfält föreslås för användning till monokulär SLAM. OpenCVs kameramodell för fisheyelinser korrigeras för att kunna hantera vinklar större än 90 grader och integrerades i ORB-SLAM2. Vi visar sedan med hjälp av fisheyemodellen, samt också MultiCol-SLAM, fördelarna med att använda vidvinkellinser för SLAM. Den prestanda gällande tracking som tillhandahålls med en fisheyelins vid snabba rotationer överträffar den för konventionella perspektivkameror. Vi hävdar att så länge som kameramodellen är tillräcklig noggrann, kommer den fullständiga bild av miljön som fås med vidvinkellins att göra monokulär SLAM robust för rotation.

La tâche consistant à tirer de l'information des images d'une caméra au cours du temps pour cartographier l'environnement et se localiser à l'intérieur de celui-ci, est appelée Localisation et Cartographie Simultanée ou SLAM.Développée à la fois par les communautés scientifiques de robotique et vision par ordinateur les applications sont multiples. Des robots bénéficient de cette capacité en gagnant en autonomie. Ces dernières années, des résultats impressionnants ont été obtenus pour des applications à des moyens de transport autonomes.Une autre champ d'application est la réalité augmenté. La localisation donnée par le SLAM offre la possibilité d'obtenir un rendu des éléments virtuels en cohérence avec les mouvements de l'utilisateur. Ainsi le cinéma, les jeux vidéos, le tourisme peuvent bénéficier de techniques SLAM. L'assistance aux travailleurs effectuant des tâches de précision ou répétitives compte également parmi les champs d'application du SLAM. Dans le cadre de cette thèse nous nous sommes intéressés au SLAM dans une optique d'applications réalistes de réalité augmentée. Bien que le sujet ait été beaucoup exploré et que d'intéressants résultats aient été obtenus, la tâche n'est toujours pas parfaitement résolue. Le problème du SLAM est un sujet de recherche ouvert, aussi bien sur des aspects spatiaux (dérive, fermeture de boucle) que temporels (temps de traitement). Dans le cadre du SLAM monoculaire nous avons surtout adressé le problème de la dérive. Puis nous nous sommes intéressés au SLAM multi-capteurs, afin d'adresser le problème des mouvements de rotation problématiques dans le cas monoculaire, et celui de la complexité calculatoire
Gathering informations from the images of a camera, over time, in order to map the environment and localize the camera in it, is a task refered to as Simultaneous Localization and Mapping, or SLAM. Developped both by the robotics and computer vision scientific communities, its applications are many. Robots gain autonomy from this ability. Quite recently, impressive results have been obtained in applications to autonomous transportation vehicles. Another field of application is augmented reality. The localization offered by SLAM enables us to display virtual objects in a consistent way a user movements. Thus, cinema, video games, tourisme applications can benefit from SLAM methods. Visual aids to workers performing complex or repetetive tasks is also an interesting application of SLAM methods. During this PhD thesis, we took interest in SLAM with the idea of realistic augmented reality applications in mind. Though the topic has been extensively explored and many impressive results obtained, the task isn't completely solved. The problem is still an open one, regarding spatial facets (drift, loop closure) as well as temporal (processing time). As part of our monocular SLAM explorations, we mainly studied the drift issue. We then explored multisensor SLAM, both as a mean to handle problematical rotational movements for the monocular setup and as mean to reduce the substantial processing times needed to solve the problem

Simultaneous Localization and Mapping (SLAM) do not result in consistent maps of large areas because of gradual increase of the uncertainty for long term missions. In addition, as the size of the map grows the computational cost increases, making SLAM solutions unsuitable for on-line applications. This thesis surveys SLAM approaches paying special attention to those approaches aimed to work on large scenarios. Special focus is given to existing underwater SLAM applications. A technique based on using independent local maps together with a global stochastic map is presented. This technique is called Selective Submap Joining SLAM (SSJS). A global map contains relative transformations between local maps, which are updated once a new loop is detected. Maps sharing several features are fused, maintaining the correlation between landmarks and vehicle's pose. The use of local maps reduces computational costs and improves map consistency as compared to state of the art techniques.
Els algoritmes de localització i creació de mapes simultàniament (Simultaneous Localization and Mapping - SLAM) no produeixen mapes correctes de grans àrees a causa de l'augment gradual de la incertesa en les missions de llarga durada. El cost de computació augmenta a mesura que el mapa creix. Aquesta tesi presenta un estudi de les tècniques de SLAM en entorns grans. També s'estudien aquells treballs centrats en ambients submarins. Es proposa una nova tècnica basada en l'ús de submapes independents i un mapa estocàstic global. Aquesta tècnica s'ha anomenat Unió Selectiva de Submapes en SLAM (SSJS). El mapa global conté les transformacions relatives entre mapes, que s'actualitzen en revisitar zones conegudes. Així doncs, els submapes que comparteixen informació es fusionen, mantenint les correlacions entre el vehicle i les fites. L'ús de submapes redueix el cost de càlcul i millora la consistència del mapa en comparació a les tècniques existents.

Questo progetto impiega le nuvole di punti estratte da un algoritmo SLAM per andare a ottenere dei miglioramenti nel calcolo di depth map da sistema monoculare. Il progetto è diviso in 2 sezioni: nella prima vengono effettuati dei test sugli algoritmi SLAM per verificare che siano dei candidati accettabili per migliorare le depth da mono; inoltre viene presentato SlamPy, una piattaforma che permette di eseguire gli stessi test su differenti algoritmi SLAM ottenendo i risultati formattati nello stesso modo. Nella seconda parte vengono applicati questi punti ad una rete sviluppata per la depth completion che permette l'utilizzo di aiuti.

This thesis reports research on mapping, path planning, and autonomous exploration. These are classical problems in robotics, typically studied independently, and here we link such problems by framing them within a common SLAM approach, adopting Pose SLAM as the basic state estimation machinery. The main contribution of this thesis is an approach that allows a mobile robot to plan a path using the map it builds with Pose SLAM and to select the appropriate actions to autonomously construct this map. Pose SLAM is the variant of SLAM where only the robot trajectory is estimated and where landmarks are only used to produce relative constraints between robot poses. In Pose SLAM, observations come in the form of relative-motion measurements between robot poses. With regards to extending the original Pose SLAM formulation, this thesis studies the computation of such measurements when they are obtained with stereo cameras and develops the appropriate noise propagation models for such case. Furthermore, the initial formulation of Pose SLAM assumes poses in SE(2) and in this thesis we extend this formulation to SE(3), parameterizing rotations either with Euler angles and quaternions. We also introduce a loop closure test that exploits the information from the filter using an independent measure of information content between poses. In the application domain, we present a technique to process the 3D volumetric maps obtained with this SLAM methodology, but with laser range scanning as the sensor modality, to derive traversability maps. Aside from these extensions to Pose SLAM, the core contribution of the thesis is an approach for path planning that exploits the modeled uncertainties in Pose SLAM to search for the path in the pose graph with the lowest accumulated robot pose uncertainty, i.e., the path that allows the robot to navigate to a given goal with the least probability of becoming lost. An added advantage of the proposed path planning approach is that since Pose SLAM is agnostic with respect to the sensor modalities used, it can be used in different environments and with different robots, and since the original pose graph may come from a previous mapping session, the paths stored in the map already satisfy constraints not easy modeled in the robot controller, such as the existence of restricted regions, or the right of way along paths. The proposed path planning methodology has been extensively tested both in simulation and with a real outdoor robot. Our path planning approach is adequate for scenarios where a robot is initially guided during map construction, but autonomous during execution. For other scenarios in which more autonomy is required, the robot should be able to explore the environment without any supervision. The second core contribution of this thesis is an autonomous exploration method that complements the aforementioned path planning strategy. The method selects the appropriate actions to drive the robot so as to maximize coverage and at the same time minimize localization and map uncertainties. An occupancy grid is maintained for the sole purpose of guaranteeing coverage. A significant advantage of the method is that since the grid is only computed to hypothesize entropy reduction of candidate map posteriors, it can be computed at a very coarse resolution since it is not used to maintain neither the robot localization estimate, nor the structure of the environment. Our technique evaluates two types of actions: exploratory actions and place revisiting actions. Action decisions are made based on entropy reduction estimates. By maintaining a Pose SLAM estimate at run time, the technique allows to replan trajectories online should significant change in the Pose SLAM estimate be detected. The proposed exploration strategy was tested in a common publicly available dataset comparing favorably against frontier based exploration

Simultaneous Localisation And Mapping SLAM-problemet är ett robotikproblem som består av att låta en robot kartlägga ett tidigare okänt område, och samtidigt lokalisera sig i den skapade kartan. Det här exjobbet presenterar ett försök till en lösning på SLAM-problemet som fungerar i konstant tid i en urban miljö. En sådan lösning måste hantera en datamängd som ständigt ökar, utan att beräkningskomplexiteten ökar signifikant.

Ett informationsfilter på fördröjd tillståndsform används för estimering av robotens trajektoria, och kamera och laseravståndssensorer används för att samla spatial information om omgivningarna längs färdvägen. Två olika metoder för att detektera loopslutningar föreslås. Den första är bildbaserad och använder Tree of Words för jämförelse av bilder. Den andra metoden är laserbaserad och använder en tränad klassificerare för att jämföra laserscans. När två posar, position och riktning, kopplats samman i en loopslutning beräknas den relativa posen med laserscansinriktning med hjälp av en kombination av Conditional Random Field-Match och Iterative Closest Point.

Experiment visar att både bild- och laserscansbaserad loopslutningsdetektion fungerar bra i stadsmiljö, och resulterar i good estimering av kartan såväl som robotens trajektoria.

In robotics, the Simultaneous Localisation And Mapping SLAM problem consists of letting a robot map a previously unknown environment, while simultaneously localising the robot in the same map. In this thesis, an attempt to solve the SLAM problem in constant time in a complex environment, such as a suburban area, is made. Such a solution must handle increasing amounts of data without significant increase in computation time.

A delayed state information filter is used to estimate the robot's trajectory, and camera and laser range sensors are used to acquire spatial information about the environment along the trajectory. Two approaches to loop closure detection are proposed. The first is image based using Tree of Words for image comparison. The second is laser based using a trained classifier for laser scan comparison. The relative pose, the difference in position and heading, of two poses matched in loop closure is calculated with laser scan alignment using a combination of Conditional Random Field-Match and Iterative Closest Point.

Experiments show that both image and laser based loop closure detection works well in a suburban area, and results in good estimation of the map as well as the robot's trajectory.

I det svenska samhället tillför diffusa och direkta föroreningskällor metaller till avloppsreningsverk, metaller som sedan hamnar i avloppsslam och recipient. Avloppsslam innehåller höga halter av växtnäringsämnen som bör återföras till jordbruksmark, men om detta ska ske får inte metallhalterna i slammet vara för höga. REVAQ är ett certifieringssystem vars syfte är att utveckla och systematisera avloppsreningsverkens uppströmsarbete och därmed möjliggöra en återföring av det växtnäringsrika slammet till jordbruksmark. Flera REVAQ-certifierade avloppsreningsverk prioriterar antimon, guld, kadmium, silver och vismut i uppströmsarbete på grund av att ackumulationshastigheten för dessa spårelement är hög i jordbruksmark som gödslas med avloppsreningsverkens slam. Source Finder (SoFi) är ett verktyg som kan användas vid uppströmsarbete för att kvantifiera identifierade källor till utsläpp av kadmium, koppar, krom, kvicksilver och zink. Syftet med examensarbetet var att vidareutveckla och anpassa verktyget SoFi till att göra beräkningar på spårelementen antimon, guld, silver och vismut samt att utveckla källan hushåll så att emissioner inom hushåll kartläggs för spårelementen och kadmium. Källor till spårelementen i avloppssystem identifierades genom litteratur och schablonvärden bestämdes för de källor det var möjligt. För att testa verktyget och kontrollera dess säkerhet utfördes en fallstudie över Käppalaverkets upptagningsområde samt en resultatkontroll. Resultaten visade att verktyget uppskattar mängden kadmium som inkommer till avloppsreningsverk bra och att inkommande mängder av antimon, silver och vismut underskattas stort. För antimon, silver och vismut var det inte möjligt att kvantifiera alla identifierade källor på grund av att kunskapen om emissioner av spårelementen är bristfällig. För guld kunde inga emissioner kvantifieras och därför beräknades inte spårelementet i verktyget. På grund av att alla stora källor inte har kvantifierats går det inte att avgöra huruvida identifieringen av källor har lyckats. Verktyget kan användas vid uppströmsarbete redan idag men genom att kvantifiera flera av de identifierade källorna kan verktyget göra större nytta och för att möjliggöra detta krävs nya studier.
Metals are transported to wastewater treatment plants (WWTPs) from diffuse and point sources in the Swedish society, these metals will end up in sewage sludge or receiving water. Sewage sludge contains a lot of plant nutrients, like phosphorus and nitrogen, which should be returned to arable land. But sludge also contains metals and if the metal content is too high the sewage sludge cannot be returned to the arable land. REVAQ is a certification system and the objective with it is to develop and systematize the WWTPs work to improve the wastewater and by that enable the return of plant nutrients to arable land. The accumulation rate of antimony, gold, cadmium, silver and bismuth are high in arable land fertilized by sewage sludge and because of that these trace elements are prioritized in WWTPs, certified by REVAQ, work to improve wastewater. Source Finder (SoFi) is a tool that can be used by the WWTPs to quantify emissions of cadmium, copper, chrome, mercury and zink from identified sources. The objectives of this master thesis were to develop and adapt the tool SoFi to estimate the trace elements antimony, gold, silver and bismuth and to develop the source household by survey the emissions of the trace elements and cadmium in it. Sources of the trace elements in the sewage system were identified by literature and emission coefficients were compiled for those sources that were possible. The new version of Source Finder was tested in the municipal WWTP Käppala and an estimation of the reasonableness of the results was made. The results showed that a good estimation of cadmium flow is made by the tool and that antimony, silver and bismuth flows are underestimated. It was not possible to determine emission coefficients for all identified sources to antimony, silver and bismuth because of the lack of knowledge about these trace elements. No emissions of gold could be quantified and the trace element was therefore not calculated by the tool. Since all identified sources have not been quantified it is not possible to determine whether all sources have been identified or not. The tool is ready to be used by the WWTPs in their work to improve the wastewater, though by quantifying further identified sources the tool will be of better use. To make that possible new studies have to be done.
Flödesanalys av spårelement från källa till slam

Lotsbroverket is the largest wastewater treatment plant on the Aland Islands and it isdesigned for handling wastewater from approximately 30 000 persons. In 2011,Lotsbroverket produced about 2800 m3 of dewatered sludge. The sewage sludge that isproduced is transported to a contractor where it is processed to eventually be used e.g.in the construction of green space. This study aims to investigate available applicationoptions in terms of the sewage sludge that is produced in Lotsbroverket. The main aimis to study the feasibility of using the produced sewage sludge as a fertilizer in theagriculture of the Aland Islands.The sludge already fulfills limit values for heavy metals in accordance with the Act"The Aland Government´s directive on the use of sewage sludge in agriculture." Inorder to clarify the sludge content of pharmaceutical and organic substances it isrequired that the substances are identified and a risk assessment is performed. In theliterature it is found that the risk of human exposure to these substances is low if thesludge is treated appropriately. Suggested appropriate treatment of the plant's sludge isthermophilic digestion whereby also pathogens are killed.The soil of the Aland Islands has a high content of phosphorus. Since 1995 there is anenvironmental program to which currently 95% of the island's farmers are connected.The program controls the use of fertilizers i.e. by setting maximum permitted levels ofphosphorus. Since sewage sludge contains relatively much phosphorus it may be alimitation of the use of sewage sludge on agricultural land of the Aland Island. That iswhy it would be suitable to use the sludge with a different fertilizer in order to obtainthe proper fertilizing properties.In Europe, the use of sewage sludge in agriculture is relatively widespread. Severalcountries have less strict laws regarding the sludge content than the Aland Islands.Because large amounts of fruits and vegetables annually are imported into the island,there is reason to believe that the population already consumes products grown onsludge treated soils. Several of the farmers on the Aland Islands are currently scepticalin terms of using sewage sludge in agriculture, mainly due to uncertainties in the sludgecontent. Regarding the certification of Lotsbroverket in accordance with the Swedishcertification system REVAQ no barriers have been found. To ensure that a certificationis possible, however, further investigations are required.

Livsformerna på jorden delas systematiskt in i de tre domänerna bakterier, arkéer och eukaryoter. Arkéer är de mikroorganismer som lever i extrema miljöer såsom hetvattenkällor, sjöar med hög salthalt och i miljöer med extrema pH-värden. De kan existera i miljöer där inga andra organismer överlever men förekommer även rikligt överallt runtomkring oss, exempelvis i människans mage och som normalflora i munnen. Vissa bakterier och arkéer har genen för enzymet ammoniak monooxygenas (AMO). Detta enzym spelar en viktig roll vid rening av avloppsvatten genom att oxidera ammonium till nitrit. Syftet med detta examensarbete har varit att detektera arkéer i prover av aktivt slam vilket gjordes genom att optimera en Polymerase Chain Reaction (PCR) baserad metod. Först pelleterades slamproverna via centrifugering för att kunna preparera DNA. Detta DNA användes som templat för optimering av PCR med specifika primers för AMO genen hos arkéer. De PCR produkter som erhölls från det optimerade programmet klonades och transformerades in i Escherichia coli. Därefter sekvenserades PCR produkterna för att identifiera vilka ammonium oxiderande arkéer som fanns i proverna. De amplifierade gensekvenserna visade god överensstämmelse med den förväntade nukleotidsekvensen för arkéer. Samtliga gensekvenser passade bäst in på icke odlingsbara arkéer enligt databasen BLAST. Genom att välja bort icke odlingsbara arkéer i sökningen kunde arkéen Nitrosopumilus maritimus identifieras, vilket är en av få odlade arkéer med AMO-genen sekvens bestämd.