Thèses sur le sujet « Robotics, software architecture, SLAM »

The goal of autonomous mobile robotics is to build physical systems that can interact with environments not specifically structured for this purpose. Even if the applications that might exploit autonomous mobile robots are widespread, current technologies are still immature at satisfying the growing requests. For this reason, robot navigation constitutes one of the major trends in the current research on robotics. A precondition for a mobile robot to be autonomous is the ability to self-localise inside an environment. This precondition is dificult to satisfy when the robot does not exploit a map of the environment to localise itself. Current research investigates methods for map learning, based on the detection of natural features. These methods should allow a robot to self-localise inside the environment it is exploring, and contemporarily to build an incremental representation of the same environment. Research on these methods is still in progress. This is due to the fact that the problem they face is hard because of the following paradox: position estimation needs a model of the environment, and world modelling needs the robot position. \Which come first, the chicken or the egg?" Current research answers the question by proposing solution based upon the simultaneity of the activities. These kinds of approach are known under the SLAM (Simultaneous Localisation And Mapping) acronym, and support the idea that the two activities should be performed together. The aim of this work is to propose a novel approach based upon the concurrence of the two activities. This approach, named CLAM (Concurrent Localisation And Mapping), is founded upon the conjecture that a proper separation of concerns may help in breaking the loop of the \chicken and egg" problem. Localisation and Modelling, acting on different time scales, are mostly independent each other. Sometimes a synchronisation is needed, but controlled by an external and suitable strategy. We consider the CLAM system a time-sensitive one since it has to perform a number of different activities with multiple, dynamic, and interdependent temporal requirements. Furthermore, a CLAM system must be able to dynamically change the activities temporal requirements. To fulfill the goal we have to define and implement a general framework for the construction of time-sensitive systems. The framework, named Real-Time Performers, is composed by a reference architecture and a working implementation providing software engineers a consistent set of software modules to build time-sensitive systems. The architecture is based upon a novel methodology based on computational reflection, this methodology models the temporal behaviour of the computational system with a set of suitable architectural abstractions, reifying time related aspects of the system itself. The final result of this work consists in a real implementation of a system supporting the exploration activity of a robot equipped with an odometric system (for positioning) and a trinocular stereo system (for environment perception). CLAM principles and Real-Time Performers architecture have driven the design of this system. Finally, the resulting system has been developed exploiting Real-Time Performers framework.

La robotique mobile est un domaine en plein essor. L'un des domaines de recherche consiste à permettre à un robot de cartographier son environnement tout en se localisant dans l'espace. Les techniques couramment employées de SLAM (Simultaneous Localization And Mapping) restent généralement coûteuses en termes de puissance de calcul. La tendance actuelle vers la miniaturisation des systèmes impose de restreindre les ressources embarquées. L'ensemble de ces constatations nous ont guidés vers l'intégration d'algorithmes de SLAM sur des architectures adéquates dédiées pour l'embarqué.Les premiers travaux ont consisté à définir une architecture permettant à un robot mobile de se localiser. Cette architecture doit respecter certaines contraintes, notamment celle du temps réel, des dimensions réduites et de la faible consommation énergétique.L'implantation optimisée d'un algorithme (EKF-SLAM), en utilisant au mieux les spécificités architecturales du système (capacités des processeurs, implantation multi-cœurs, calcul vectoriel ou parallélisation sur architecture hétérogène), a permis de démontrer la possibilité de concevoir des systèmes embarqués pour les applications SLAM dans un contexte d'adéquation algorithme architecture. Une seconde approche a été explorée ayant pour objectif la définition d'un système à base d'une architecture reconfigurable (à base de FPGA) permettant la conception d'une architecture fortement parallèle dédiée au SLAM. L'architecture définie a été évaluée en utilisant une méthodologie HIL (Hardware in the Loop).Les principaux algorithmes de SLAM sont conçus autour de la théorie des probabilités, ils ne garantissent en aucun cas les résultats de localisation. Un algorithme de SLAM basé sur la théorie ensembliste a été défini garantissant l'ensemble des résultats obtenus. Plusieurs améliorations algorithmiques sont ensuite proposées. Une comparaison avec les algorithmes probabilistes a mis en avant la robustesse de l'approche ensembliste.Ces travaux de thèse mettent en avant deux contributions principales. La première consiste à affirmer l'importance d'une conception algorithme-architecture pour résoudre la problématique du SLAM. La seconde est la définition d'une méthode ensembliste permettant de garantir les résultats de localisation et de cartographie.

Doutoramento em Ciências da Computação
When developing software for autonomous mobile robots, one has to inevitably tackle some kind of perception. Moreover, when dealing with agents that possess some level of reasoning for executing their actions, there is the need to model the environment and the robot internal state in a way that it represents the scenario in which the robot operates. Inserted in the ATRI group, part of the IEETA research unit at Aveiro University, this work uses two of the projects of the group as test bed, particularly in the scenario of robotic soccer with real robots. With the main objective of developing algorithms for sensor and information fusion that could be used e ectively on these teams, several state of the art approaches were studied, implemented and adapted to each of the robot types. Within the MSL RoboCup team CAMBADA, the main focus was the perception of ball and obstacles, with the creation of models capable of providing extended information so that the reasoning of the robot can be ever more e ective. To achieve it, several methodologies were analyzed, implemented, compared and improved. Concerning the ball, an analysis of ltering methodologies for stabilization of its position and estimation of its velocity was performed. Also, with the goal keeper in mind, work has been done to provide it with information of aerial balls. As for obstacles, a new de nition of the way they are perceived by the vision and the type of information provided was created, as well as a methodology for identifying which of the obstacles are team mates. Also, a tracking algorithm was developed, which ultimately assigned each of the obstacles a unique identi er. Associated with the improvement of the obstacles perception, a new algorithm of estimating reactive obstacle avoidance was created. In the context of the SPL RoboCup team Portuguese Team, besides the inevitable adaptation of many of the algorithms already developed for sensor and information fusion and considering that it was recently created, the objective was to create a sustainable software architecture that could be the base for future modular development. The software architecture created is based on a series of di erent processes and the means of communication among them. All processes were created or adapted for the new architecture and a base set of roles and behaviors was de ned during this work to achieve a base functional framework. In terms of perception, the main focus was to de ne a projection model and camera pose extraction that could provide information in metric coordinates. The second main objective was to adapt the CAMBADA localization algorithm to work on the NAO robots, considering all the limitations it presents when comparing to the MSL team, especially in terms of computational resources. A set of support tools were developed or improved in order to support the test and development in both teams. In general, the work developed during this thesis improved the performance of the teams during play and also the e ectiveness of the developers team when in development and test phases.
Durante o desenvolvimento de software para robôs autónomos móveis, e inevitavelmente necessário lidar com algum tipo de perceção. Al em disso, ao lidar com agentes que possuem algum tipo de raciocínio para executar as suas ações, há a necessidade de modelar o ambiente e o estado interno do robô de forma a representar o cenário onde o robô opera. Inserido no grupo ATRI, integrado na unidade de investigação IEETA da Universidade de Aveiro, este trabalho usa dois dos projetos do grupo como plataformas de teste, particularmente no cenário de futebol robótico com robôs reais. Com o principal objetivo de desenvolver algoritmos para fusão sensorial e de informação que possam ser usados eficazmente nestas equipas, v arias abordagens de estado da arte foram estudadas, implementadas e adaptadas para cada tipo de robôs. No âmbito da equipa de RoboCup MSL, CAMBADA, o principal foco foi a perceção da bola e obstáculos, com a criação de modelos capazes de providenciar informação estendida para que o raciocino do robô possa ser cada vez mais eficaz. Para o alcançar, v arias metodologias foram analisadas, implementadas, comparadas e melhoradas. Em relação a bola, foi efetuada uma análise de metodologias de filtragem para estabilização da sua posição e estimação da sua velocidade. Tendo o guarda-redes em mente, foi também realizado trabalho para providenciar informação de bolas no ar. Quanto aos obstáculos, foi criada uma nova definição para a forma como são detetados pela visão e para o tipo de informação fornecida, bem como uma metodologia para identificar quais dos obstáculos são colegas de equipa. Além disso foi desenvolvido um algoritmo de rastreamento que, no final, atribui um identicador único a cada obstáculo. Associado a melhoria na perceção dos obstáculos foi criado um novo algoritmo para realizar desvio reativo de obstáculos. No contexto da equipa de RoboCup SPL, Portuguese Team, al em da inevitável adaptação de vários dos algoritmos j a desenvolvidos para fusão sensorial e de informação, tendo em conta que foi recentemente criada, o objetivo foi criar uma arquitetura sustentável de software que possa ser a base para futuro desenvolvimento modular. A arquitetura de software criada e baseada numa série de processos diferentes e métodos de comunicação entre eles. Todos os processos foram criados ou adaptados para a nova arquitetura e um conjunto base de papeis e comportamentos foi definido para obter uma framework funcional base. Em termos de perceção, o principal foco foi a definição de um modelo de projeção e extração de pose da câmara que consiga providenciar informação em coordenadas métricas. O segundo objetivo principal era adaptar o algoritmo de localização da CAMBADA para funcionar nos robôs NAO, considerando todas as limitações apresentadas quando comparando com a equipa MSL, principalmente em termos de recursos computacionais. Um conjunto de ferramentas de suporte foram desenvolvidas ou melhoradas para auxiliar o teste e desenvolvimento em ambas as equipas. Em geral, o trabalho desenvolvido durante esta tese melhorou o desempenho da equipas durante os jogos e também a eficácia da equipa de programação durante as fases de desenvolvimento e teste.

Cette thèse se focalise sur des tâches de perceptions pour des drones à voilures fixes (UAV). Lorsque la perception est la finalité, un bon modèle d'environnement couplé à la capacité de prédire l'impact de futures observations sur celui-ci est crucial. La perception active traite de l'intégration forte entre modèles de perception et processus de raisonnement, permettant au robot d'acquérir des informations pertinentes à propos du statut de la mission et de replanifier sa trajectoire de mesure en réaction à des évènements et résultats imprévisibles. Ce manuscrit décrit deux approches pour des tâches de perception active, dans deux scénarios radicalement différents. Le premier est celui de la cartographie des phénomènes météorologiques de petite échelle et fortement dynamiques, en particulier de nuages de type cumulus. L'approche présentée utilise la régression par processus Gaussien pour construire un modèle d'environnement, les hyper-paramètres étant appris en ligne. Des métriques de gain d'information sont introduites pour évaluer la qualité de futures trajectoires d'observation. Un algorithme de planification stochastique est utilisé pour optimiser une fonction d'utilité équilibrant maximisation du gain d'information avec des buts de minimisation du coût énergétique. Dans le second scénario, un UAV cartographie des champs de grandes cultures pour les besoins de l'agriculture de précision. Utilisant le résultat d'un algorithme de localisation et cartographie simultanée (SLAM), une approche nouvelle pour la construction d'un modèle d'erreurs relatives est proposée. Ce modèle est appris à partir d'attributs provenant des structures de données du SLAM, ainsi que de la topologie sous-jacente du graphe de covisibilité formé par les observations. Tous les développement ont été testés en simulation. Se focalisant sur la problématique de gestion de l'avancement tu temps et de la synchronisation de simulateurs hétérogènes dans une architecture distribuée, une solution originale basée sur une architecture décentralisée est proposée
This thesis focuses on perception tasks for an unmanned aerial vehicle (UAV). When sensing is the finality, having a good environment model as well as being capable of predicting the impacts of future observations is crucial. Active perception deals with integrating tightly perception models in the reasoning process, enabling the robot to gain knowledge about the status of its mission and to replan its sensing trajectory to react to unforeseen events and results. This manuscript describes two approaches for active perception tasks, in two radically different settings. The first one deals with mapping highly dynamic and small scale meteorological phenomena such as cumulus clouds. The presented approach uses Gaussian Process Regression to build environment models, learning its hyperparameters online. Normalized marginal information metrics are introduced to compute the quality of future observation trajectories. A stochastic planning algorithm is used to optimize an utility measure balancing maximization of theses metrics with energetic minimization goals. The second setting revolves around mapping crop fields for precision agriculture purposes. Using the output of a monocular graph Simultaneous Localization and Mapping (SLAM) algorithm, a novel approach to building a relative error model is proposed. This model is learned both from features extracted from the SLAM algorithm’s data structures, as well as the underlying topology of the covisibility graph of the observations. All developments have been tested using realistic, distributed simulations. An analysis of the simulation issue in robotics is proposed. Focusing on the problem of managing time advancement of multiple interconnected simulators, a novel solution based on a decentralized scheme is presented

Software used in medical settings operate in complex and variable environments. Programs need to integrate well not only with their electrical and mechanical components, but also within the socio-technological setting they participate in. In this Master's Thesis, a modular software architecture for controlling surgical robot systems within magnetic resonance scanners is designed and implemented. The C++ program runs on a sbRIO 9651 real-time operating system and an object oriented design is taken. Robot kinematics and controls are put into effect in software and validated. Communication with up to ten daughter cards occurs via SPI and external information is exchanged via OpenIGTLink. A web-based engineering console made with ReactJS is also constructed to provide a visual interface for actuating motor axes and executing robot functionality. Documentation of the code is provided and the program was validated quantitatively with software tests and qualitatively through experimentation in MRI suites.

Este trabalho apresenta uma arquitetura de hardware, baseada em FPGA (Field-Programmable Gate Array) e com multi-câmeras, para o problema de localização e mapeamento simultâneos - SLAM (Simultaneous Localization And Mapping) aplicada a sistemas robóticos embarcados. A arquitetura é composta por módulos de hardware altamente especializados para a localização do robô e para geração do mapa do ambiente de navegação em tempo real com features extraídas de imagens obtidas diretamente de câmeras CMOS a uma velocidade de 30 frames por segundo. O sistema é totalmente embarcado em FPGA e apresenta desempenho superior em, pelo menos, uma ordem de magnitude em relaçãoo às implementações em software processadas por computadores pessoais de última geração. Esse desempenho deve-se à exploração do paralelismo em hardware junto com o processamento em pipeline e às otimizações realizadas nos algoritmos. As principais contribuições deste trabalho são as arquiteturas para o filtro de Kalman estendido - EKF (Extended Kalman Filter) e para a detecção de features baseada no algoritmo SIFT (Scale Invariant Feature Transform). A complexidade para a implementaçãoo deste trabalho pode ser considerada alta, uma vez que envolve uma grande quantidade de operações aritméticas e trigonométricas em ponto utuante e ponto fixo, um intenso processamento de imagens para extração de features e verificação de sua estabilidade e o desenvolvimento de um sistema de aquisição de imagens para quatro câmeras CMOS em tempo real. Adicionalmente, foram criadas interfaces de comunicação para o software e o hardware embarcados no FPGA e para o controle e leitura dos sensores do robô móvel. Além dos detalhes e resultados da implementação, neste trabalho são apresentados os conceitos básicos de mapeamento e o estado da arte dos algoritmos SLAM com visão monocular e estéreo
This work presents a hardware architecture for the Simultaneous Localization And Mapping (SLAM) problem applied to embedded robots. This architecture, which is based on FPGA and multi-cameras, is composed by highly specialized blocks for robot localization and feature-based map building in real time from images read directly from CMOS cameras at 30 frames per second. The system is completely embedded on an FPGA and its performance is at least one order of magnitude better than a high end PC-based implementation. This result is achieved by investigating the impact of several hardwareorientated optimizations on performance and by exploiting hardware parallelism along with pipeline processing. The main contributions of this work are the architectures for the Extended Kalman Filter (EKF) and for the feature detection system based on the SIFT (Scale Invariant Feature Transform). The complexity to implement this work can be considered high, as it involves a significant number of arithmetic and trigonometric operations in oating and fixed-point format, an intensive image processing for feature detection and stability checking, and the development of an image acquisition system from four CMOS cameras in real time. In addition, communication interfaces were created to integrate software and hardware embedded on FPGA and to control the mobile robot base and to read its sensors. Finally, besides the implementation details and the results, this work also presents basic concepts about mapping and state-of-the-art algorithms for SLAM with monocular and stereo vision.

La plupart des applications robotiques, telles que les véhicules autonomes, sont développées à partir d’une page blanche avec quelques rares réutilisations de conceptions ou de codes issus d’anciens projets équivalents. Qui plus est, les systèmes robotiques deviennent de plus en plus critiques, dans la mesure où ils sont déployés dans des environnements peu structurés, et centrés sur l’humain. Ces systèmes à fort contenu logiciel qui utilisent des composants distribués et hétérogènes interagissent dans un environnement dynamique, et incertain. Or, il s’agit là d’étapes indispensables pour la mise en place de méthodes d’évaluation extensibles, ainsi que pour permettre la réutilisation de composants logiciels pré-existants. Le développement de structures logicielles et d’outils de conception d’architectures, orientés pour la robotique, coûte cher en termes de temps et d’effort, et l’absence d’une approche systématique pourrait conduire à la production de conceptions adhoc, peu flexibles et peu réutilisables. Faire de la meta-structure de l’architecture un point de convergence offre de nouvelles possibilités en termes d’interopérabilité, et de partage de la connaissance, au sein des communautés dédiées à la mise en place d’architectures et de structures. Nous suivons cette direction, en proposant un modèle commun, et en fournissant une approche méthodologique systématique aidant à spécifier les différents aspects du développement d’architectures logicielles, et leurs relations au sein d’une structure partagée
Most innovative applications having robotic capabilities like self-driving cars are developed from scratch with little reuse of design or code artifacts from previous similar projects. As a result, work at times is duplicated adding time and economic costs. Absence of integrated tools is the real barrier that exists between early adopters of standardization efforts and early majority of research and industrial community. These software intensive systems are composed of distributed, heterogeneous software components interacting in a highly dynamic, uncertain environment. However, no significant systematic software development process is followed in robotics research. The process of developing robotic software frameworks and tools for designing robotic architectures is expensive both in terms of time and effort, and absence of systematic approach may result in ad hoc designs that are not flexible and reusable. Making architecture meta-framework a point of conformance opens new possibilities for interoperability and knowledge sharing in the architecture and framework communities. We tried to make a step in this direction by proposing a common model and by providing a systematic methodological approach that helps in specifying different aspects of software architecture development and their interplay in a framework

Manufacturing, under the umbrella of the latest industrial revolution, has gone through enormous changes in the last decades to then later evolve in what we know now as smart manufacturing. Different companies and entities have developed their own versions of architectures for intelligentand digitalized manufacturing systems. Ideating a exible and safe architecture is one of the first steps towards a system that intends to be applicable in different environments, regardless of the vast variety of possibilities available. For this purpose, the following thesis presents an investigation on the state-of-the-art solutions of the most recent digitalized cloud-based system architectures in the domain of discreet manufacturing. Based on an initial system architecture conceived from the company ABB, an evaluation of this architecture was conducted, by taking in consideration the existing systematical approaches to the digitalization of this industry. In the following thesis work, we investigate, describe and evaluate the limitations and strengths of the most recent and known architectural approaches to cloud robotics. Finally, a few key remarks are made towards ABB's initial solution but also to the industry in general.
PADME

La Robotique a connu une évolution remarquable au cours des dernières années, couplée à un intérêt croissant de la société pour ce domaine. Les robots ne sont plus fabriqués exclusivement pour effectuer des tâches répétitives dans les usines, mais ils sont aussi créés pour collaborer avec les humains dans plusieurs domaines d'application d'importance. Les systèmes robotiques qui contrôlent ces robots sont donc de plus en plus larges, complexes et difficiles à développer. Dans ce contexte, l'Architecture Orientée Services (SOA) a été identifiée comme un style d'architecture logicielle prometteur pour concevoir des systèmes robotiques de manière flexible, réutilisable et productive. Cependant, malgré le nombre considérable de Systèmes Robotiques Orientées Services (SORS) existants aujourd'hui, la plupart d'entre eux ont été développés de manière ad hoc. Le peu d'attention et le soutien limité portés à la conception d'architectures logicielles SORS peuvent non seulement masquer les avantages de l'adoption de la SOA, mais aussi réduire la qualité globale des systèmes robotiques, qui sont souvent utilisés dans des contextes de sécurité critiques. Cette thèse vise à améliorer la compréhension et la systématisation de la conception architecturale SORS. Elle décrit une taxonomie des services pour le domaine de la robotique, puis propose un processus ainsi qu'une architecture de référence afin de systématiser la conception d'architectures logicielles SORS. Les résultats obtenus dans les études d'évaluation montrent qu'à la fois le processus et l'architecture de référence peuvent avoir un impact positif sur la qualité des architectures logicielles SORS et, par conséquent, contribuent à l'amélioration des systèmes robotiques
Robotics has experienced an increasing evolution and interest from the society in recent years. Robots are no longer produced exclusively to perform repetitive tasks in factories, they have been designed to collaborate with humans in several important application domains. Robotic systems that control these robots are therefore becoming larger, more complex, and difficult to develop. In this scenario, Service-Oriented Architecture (SOA) has been investigated as a promising architectural style for the design of robotic systems in a flexible, reusable, and productive manner. Despite the existence of a considerable amount of Service-Oriented Robotic Systems (SORS), most of them have been developed in an ad hoc manner. The little attention and limited support devoted to the design of SORS software architectures may not only hamper the benefits of SOA adoption, but also reduce the overall quality of robotic systems, which are often used in safety-critical contexts. This thesis aims at improving the understanding and systematization of SORS architectural design

A robótica tem passado por uma notável evolução ao longo dos últimos anos, juntamente com um crescente interesse por parte da sociedade. Robôs não são mais exclusivamente produzidos para realizar atividades repetitivas em fábricas, eles têm sido projetados para apoiar humanos em diversos e importantes domínios de aplicação. Os sistemas robóticos utilizados para controlar tais robôs têm, portanto, se tornado maiores, mais complexos e difíceis de desenvolver. Nesse cenário, a Arquitetura Orientada a Serviços (do inglês, Service-Oriented Architecture - SOA) tem sido investigada como um promissor estilo arquitetural para o desenvolvimento de sistemas robóticos de forma mais flexível, reusável e produtiva. Embora um número considerável de Sistemas Robóticos Orientados a Serviços (do inglês, Service-Oriented Robotic Systems - SORS) já exista, grande parte deles têm sido desenvolvida de maneira ad hoc. A pouca atenção e o suporte limitado ao projeto das arquiteturas de software de SORS pode não só impedir a obtenção dos benefícios associados à adoção da SOA, mas também reduzir a qualidade dos sistemas robóticos que, frequentemente, são utilizados em contextos de segurança crítica. Essa tese tem por objetivo aprimorar o entendimento e a sistematização do projeto arquitetural de SORS. Para isso, é proposta uma taxonomia de serviços para o domínio de robótica, bem como um processo e uma arquitetura de referência para sistematizar o projeto das arquiteturas de software de SORS. Os resultados obtidos evidenciam que tanto o processo quanto a arquitetura de referência podem impactar positivamente na qualidade das arquiteturas de software de SORS e, consequentemente, contribuir para o desenvolvimento de sistemas robóticos.
Robotics has experienced an increasing evolution and interest from the society in recent years. Robots are no longer produced exclusively to perform repetitive tasks in factories, they have been designed to collaborate with humans in several important application domains. Robotic systems that control these robots are therefore becoming larger, more complex, and difficult to develop. In this scenario, Service-Oriented Architecture (SOA) has been investigated as a promising architectural style for the design of robotic systems in a exible, reusable, and productive manner. Despite the existence of a considerable amount of Service-Oriented Robotic Systems (SORS), most of them have been developed in an ad hoc manner. The little attention and limited support devoted to the design of SORS software architectures may not only hamper the benefits of SOA adoption, but also reduce the overall quality of robotic systems, which are often used in safety-critical contexts. This thesis aims at improving the understanding and systematization of SORS architectural design. It describes a taxonomy of services for the robotics domain, as well as proposes a process and a reference architecture that systematize the design of SORS software architectures. Results achieved in the evaluation studies evidence that both process and reference architecture can positively impact on the quality of SORS software architectures and, as a consequence, contribute to the development of robotic systems.

Sistemas embarcados, em especial aqueles utilizados em robótica, apresentam, em sua estrutura, uma multiplicidade de dispositivos que resultam em uma arquitetura bastante heterogênea e bem distribuída. Para auxiliar na resolução dessa complexidade inerente, este trabalho resgata os conceitos de frameworks, buscando na sua integração e modelo conceitual, desenvolver um conjunto de ferramentas que gerencia a mediação entre sistemas embarcados e demais aplicações de software, fornecendo bibliotecas e componentes reutilizáveis para aplicações na robótica. Propõe-se utilizar framework e middleware de sistemas open-source para integração entre a plataforma de software e hardware. Um dos objetivos do projeto é criar um framework multi-plataforma com diferentes tipos de serviços para o de desenvolvimento de aplicações no campo da robótica. O projeto tem como estudo de caso sistemas embarcados aplicados em robótica móvel e tecnologia assistiva.
Embedded systems, especially those used in robotics, present in its structure, a plurality of devices that result in a very heterogeneous and well distributed architecture. To help resolve this inherent complexity, the work rescues the concepts of frameworks, seeking their integration and conceptual model, develop a set of tools that manage to mediate between embedded systems and other of software applications, providing reusable libraries and components for applications in robotics. It is proposed to use framework and middleware systems open source for integration between the platform software and hardware. One of the project objectives is to create a multi-platform framework with different types of services, considering the programming and compatibility with hardware for the development of applications in the field of robotics. The project’s case study developing embedded applied in mobile and Assistive Technology robotics.

Un système robotique est un système complexe, à la fois d’un point de vue matériel et logiciel. Afin de simplifier la conception de ces machines, le développement est découpé en modules qui sont ensuite assemblés pour constituer le système complet. Cependant, la facilité de conception de ces systèmes est bien souvent contrebalancée par la complexité de leur mise en sécurité, à la fois d’un point de vue fonctionnel et temporel. Il existe des ensembles d’outils et de méthodes permettant l’étude d’ordonnançabilité d’un système logiciel à base de tâches. Ces outils permettent de vérifier qu’un système de tâches respecte ses contraintes temporelles. Cependant ces méthodes d’analyse considèrent les tâches comme des entités monolithiques, sans prendre en compte la structure interne des tâches, ce qui peut les rendre trop pessimistes et non adaptées à des applications robotiques. Cette étude consiste à prendre en compte la structure interne des tâches dans des méthodes d’analyse d’ordonnançabilité. Cette thèse montre que le découpage de tâches monolithiques permet d’améliorer la précision des analyses d’ordonnancement. De plus, les outils issus de ces travaux ont été expérimentés sur un cas d’application de robotique mobile autonome
A robot is a complex system combining hardware and software parts. In order to simplify the robot design, the whole system is split in several separated modules. However, the complexity of the functional and temporal validation to improve the safety counterweights the robot design simplicity. We can find scheduling analysis tools for task-based software. These tools are used to check and validate the schedulability of the tasks involved in a software, run on a specific hardware. However, these methods considers the tasks as monolithic entities, without taking into account their internal structure. The resulting analyses may be too much pessimistic and therefore not applicable to robotic applications. In this work, we have modeled the internal structure of the tasks as state-machines and used these state-machines into the schedulability analysis in order to improve the analysis precision. Moreover, the tools developed during this work have been tested on real robotic use-cases

Les applications sous-marines actuelles exigent la réalisation de travaux de différentes natures dans des zones de plus en plus vastes et toujours plus profondes. La conception et l'utilisation de flottilles d'AUV dans ce contexte est un véritable challenge. Les bénéfices attendus sontmultiples. Premièrement, cela doit permettre de minimiser les coûts grâce à une répartition de ceux-ci sur l'ensemble de la flottille : la perte d'un AUV ou son mauvais fonctionnement ne remettront pas en cause l'intégralité de la mission. Deuxièmement, l'utilisation d'une flottilledoit de fait permettre de réduire le temps d'exécution d'une mission grâce à la parallélisation de certaines tâches. Enfin, la réalisation d'une mission par une flottille permet de conserver le caractère spécialisé des AUV et donc d'envisager plus facilement leur réutilisation dansd'autres contextes. Cependant, les approches de coopération multi-AUV existantes sont limitées par deux principaux verrous : (1) le nombre de communications induit et (2) la gestion de l'hétérogénéité potentielle d'une flottille.L'approche que nous proposons vise à répondre à ces problématiques.L'idée principale est de combiner une approche de coopération réactive avec une approche organisationnelle. L'approche de coopération réactive permet l'échange des signaux de communication très simples. Cependant, elle ne permet de résoudre que des problèmes de coopération de nature assez restreinte qui concernent essentiellement la coordination spatiale de véhicules homogènes. La première contribution de cette thèse est l'extension de l'approche satisfaction-altruisme. Un nouveau mécanisme décisionnel réactif, capable de considérer des actions coopératives de diverses natures, est proposé. La deuxième contribution consiste à spécifier les contextes d'interactions réactives à l'aide d'une approche organisationnelle. Le modèle organisationnel Agent/Groupe/Rôle est utilisé pour avoir une représentation explicite de l'organisation de la flottille. Les concepts de groupe, mais surtout de rôle, sont employés dans l'adressage des signaux de communication et permettent la mise en oeuvre d'interactions hétérogènes avec une grande modularité. L'ensemble est intégré dans une nouvelle architecture logicielle, appelée REMORAS, destinée à équiper des véhicules sous-marins autonomes. Une validation de la faisabilité de notre approche est proposée à l'aide d'une simulation de plusieurs scénarios mettant en jeu des AUV hétérogènes
Underwater marine applications are nowadays branching into various fields covering larger and deeper zones. Performing the required tasks with the aid of AUV flotillas is a real challenge. However, the advantages of using such a new technology are numerous. Firstly, this would highly reduce the cost of the mission thanks to the distribution of this former among the various AUV: the loss of one AUV or its bad functioning will not degrade the performance of the flotilla in general. Secondly, the use of a flotilla reduces the execution time of a mission given the parallelization of certain tasks. Finally, any mission can be accomplished by the flotilla by taking into consideration the specificity of each AUV. In fact, each of these vehicles holds different characteristics rendering the global architecture heterogeneous and therefore applicable in different contexts. However, the methods concerned with multi-AUV cooperation are hindered by two main limitations: (1) the number of communications induced and (2) the management of the heterogeneity in the flotilla.The proposed approach aims at responding to these challenges. The principal idea is to combine this reactive cooperational approach with an organizational one. The reactive cooperational approach allows the exchange of simple communication signals. However, it does not help in solving the problems of cooperation that are very constrained and that mainly concern the spatial coordination of homogeneous vehicles. The first contribution in this thesis is the extension of the satisfaction-altruism approach. A new reactive decisional mechanism capable of considering the cooperative actions of various natures is proposed. The second contribution consists in specifying the context of reactive interactions based on an organizational approach. The organizational model Agent/Group/Role is used in order to have an explicit representation of the flotilla. The concepts of "group" and especially "role" are used in the attribution of the communication signals allowing the accomplishment of heterogeneous interactions with a big modularity. A new concept is therefore born and is integrated in a new software architecture called REMORA intended to equip autonomous underwater vehicles. This proposed new method has been validated through various numerical simulations in different scenarios putting at stake heterogeneous AUV

Arctic regions are thought to be more sensitive to climate change fluctuations, making weather data from these regions more valuable for climate modeling. Scientists have expressed an interest in deploying a robotic sensor network in these areas, minimizing the exposure of human researchers to the harsh environment, while allowing dense, targeted data collection to commence. For any such robotic system to be successful, a certain set of base navigational functionality must be developed. Further, these navigational algorithms must rely on the types of low-cost sensors that would be viable for use in a multi-agent system. A set of vision-based processing techniques have been proposed, which augment current robotic technologies for use in glacial terrains. Specifically, algorithms for estimating terrain traversability, robot localization, and terrain reconstruction have been developed which use data collected exclusively from a single camera and other low-cost robotic sensors. For traversability assessment, a custom algorithm was developed that uses local scale surface texture to estimate the terrain slope. Additionally, a horizon line estimation system has been proposed that is capable of coping with low-contrast, ambiguous horizons. For localization, a monocular simultaneous localization and mapping (SLAM) filter has been fused with consumer-grade GPS measurements to produce full robot pose estimates that do not drift over long traverses. Finally, a terrain reconstruction methodology has been proposed that uses a Gaussian process framework to incorporate sparse SLAM landmarks with dense slope estimates to produce a single, consistent terrain model. These algorithms have been tested within a custom glacial terrain computer simulation and against multiple data sets acquired during glacial field trials. The results of these tests indicate that vision is a viable sensing modality for autonomous glacial robotics, despite the obvious challenges presented by low-contrast glacial scenery. The findings of this work are discussed within the context of the larger arctic sensor network project, and a direction for future work is recommended.

Der vorliegende Beitrag stellt einen Entwurf für eine flexible und robuste Steuerungsarchitektur für Roboter- und Fertigungssysteme vor. Dabei wurde versucht ein offenes Konzept zu realisieren, welches einen vereinfachten Engineeringprozess ermöglicht. Hierzu wird innerhalb der Steuerung eine Trennung zwischen einem funktionellen verhaltensbasierten und einem ablauforientierten Modell vorgeschlagen. Dieser Ansatz wird durch die Verwendung von Aktionsprimitiven innerhalb einer hybriden Robotersteuerung ermöglicht. Diese garantieren durch ihre ausgeprägte Modularität eine hohe Flexibilität und Erweiterbarkeit des entstandenen Systems. Im Beitrag wird sowohl der entstandene Entwurf diskutiert als auch eine prototypische objektorientierte Implementierung vorgestellt sowie erste Ergebnisse präsentiert
This paper presents a framework for a flexible and robust control architecture for robotic systems. The design incorporates an application independent system concept which allows a simplified engineering process. For this purpose a distinction between a functional behavioural and a sequential control system model is proposed. This approach is based on the utilisation of action primitives within a hybrid control architecture. The use of these primitives affords a high level of modularity through increasing flexibility and expandability of the resulting system. In this paper the proposed framework will be discussed as well as a prototypical object-oriented implementation and first results

In der vorliegenden Arbeit wird die flexible Steuerungsarchitektur Apeca für Systeme der Robotik sowie der robotergestützten Fertigungstechnik vorgestellt. Dafür werden verschiedene Anforderungen identifiziert und innerhalb eines Entwurfs vereint. Ein Hauptaugenmerk des dabei entstandenen Konzeptes ist es, einen vereinfachten Engineeringprozess für den Steuerungsentwurf zu ermöglichen. Dieser Ansatz wird durch die Verwendung von Aktionsprimitiven ermöglicht, die in Form atomarer Systemverhalten in einer speziellen Modulhierarchie eingesetzt werden. Hierzu erfolgt innerhalb der Steuerungsarchitektur eine Trennung zwischen einem funktionsorientierten verhaltensbasierten Modell zur hierarchischen sowie funktionell parallelen Ausführung von Aktionsprimitiven und einem ablauforientierten Modell zur aufgabenabhängigen Aktivierung derselben. Mit Hilfe eines Nutzerkonzepts werden diese Modelle verschiedenen Anwendern zugeordnet. Die objektorientierte Realisierung dieses Entwurfs ermöglicht die Verwendung und Synchronisation von mehreren Teilsystemen innerhalb einer Steuerung. In der Arbeit wird sowohl der entstandene Entwurf diskutiert als auch eine prototypische Implementierung vorgestellt. Abschließend werden die Ergebnisse anhand verschiedener Demonstrationsszenarien präsentiert
In this present work, the Apeca framework, a flexible control architecture for robotic systems, is introduced. The conceptual design combines different requirements identified in miscellaneous robotic control approaches. The main focus of the resulting concept is on a simplified engineering process for the controller design. This approach is supported by the use of atomic system behaviors, the so called action primitives, in a special module hierarchy. For this purpose a distinction between a functional behavior based system model with hierarchically and also parallelly executed action primitives and a sequential control system model with a task-dependent activation of the primitives is proposed. These models are assigned to different users through a distinct user concept. An object-oriented implementation of the proposed architecture allows the utilization and synchronisation of multiple (sub-)systems within one framework. In this work the proposed framework will be discussed, a prototypical implementation will be presented and results based on different experimental scenarios will be shown

Home environments are changing as more technological devices are used to improve daily life. The growing demand for high technology in our homes means that robot integration will soon arrive. Home devices are evolving in a connected paradigm in which data flows to perform efficient home task management. Heterogeneous home robots connected in a network can establish a workflow that complements their capabilities and so increases performance within a mission execution. This work addresses the definition and requirements of a robot-group mission in the home context. The proposed solution relies on a network of smart resources, which are defined as cyber-physical systems that provide high-level service execution. Firstly, control middleware architecture is introduced as the execution base for the Smart resources. Next, the Smart resource topology and its integration within a robotic platform are addressed. Services supplied by Smart resources manage their execution through a robot behavior architecture. Robot behavior execution is hierarchically organized through a mission definition that can be established as an individual or collective approach. Environment model and interaction tasks characterize the operation capabilities of each robot within a mission. Mission goal achievement in a heterogeneous group is enhanced through the complement of the interaction capabilities of each robot. To offer a clearer explanation, a full use case is presented in which two robots cooperate to execute a mission and the previously detailed steps are evaluated. Finally, some of the obtained results are discussed as conclusions and future works is introduced.
Los entornos domésticos se encuentran sometidos a un proceso de cambio gracias al empleo de dispositivos tecnológicos que mejoran la calidad de vida de las personas. La creciente demanda de alta tecnología en los hogares señala una próxima incorporación de la robótica de servicio. Los dispositivos domésticos están evolucionando hacia un paradigma de conexión en el cual la información fluye para ofrecer una gestión más eficiente. En este entorno, robots heterogéneos conectados a la red pueden establecer un flujo de trabajo que ofreciendo nuevas soluciones y incrementando la eficiencia en la ejecución de tareas. Este trabajo aborda la definición y los requisitos necesarios para la ejecución de misiones en grupos de robots heterogéneos en entornos domésticos. La solución propuesta se apoya en una red de Smart resources, que son definidos como sistemas ciber-físicos que proporcionan servicios de alto nivel. En primer lugar, se presenta la arquitectura del middleware de control en la cual se basa la ejecución de los Smart resources. A continuación se detalla la topología de los Smart resources, así como su integración en plataformas robóticas. Los servicios proporcionados por los Smart resources gestionan su ejecución mediante una arquitectura de comportamientos para robots. La ejecución de estos comportamientos se organiza de forma jerárquica mediante la definición de una misión con un objetivo establecido de forma individual o colectiva a un grupo de robots. Dentro de una misión, las tareas de modelado e interacción con el entorno define las capacidades de operación de los robots dentro de una misión. Mediante la integración de un grupo heterogéneo de robots sus diversas capacidades son complementadas para el logro un objetivo común. A fin de caracterizar esta propuesta, los mecanismos presentados en este documento se evaluarán en detalle a lo largo de una serie experimentos en los cuales un grupo de robots heterogéneos ejecutan una misión colaborativa para alcanzar un objetivo común. Finalmente, los resultados serán discutidos a modo de conclusiones dando lugar el establecimiento de un trabajo futuro.
Els entorns domèstics es troben sotmesos a un procés de canvi gràcies a l'ocupació de dispositius tecnològics que milloren la qualitat de vida de les persones. La creixent demanda d'alta tecnologia a les llars assenyala una propera incorporació de la robòtica de servei. Els dispositius domèstics estan evolucionant cap a un paradigma de connexió en el qual la informació flueix per oferir una gestió més eficient. En aquest entorn, robots heterogenis connectats a la xarxa poden establir un flux de treball que ofereix noves solucions i incrementant l'eficiència en l'execució de tasques. Aquest treball aborda la definició i els requisits necessaris per a l'execució de missions en grups de robots heterogenis en entorns domèstics. La solució proposada es recolza en una xarxa de Smart resources, que són definits com a sistemes ciber-físics que proporcionen serveis d'alt nivell. En primer lloc, es presenta l'arquitectura del middleware de control en la qual es basa l'execució dels Smart resources. A continuació es detalla la tipologia dels Smart resources, així com la seva integració en plataformes robòtiques. Els serveis proporcionats pels Smart resources gestionen la seva execució mitjançant una arquitectura de comportaments per a robots. L'execució d'aquests comportaments s'organitza de forma jeràrquica mitjançant la definició d'una missió amb un objectiu establert de forma individual o col·lectiva a un grup de robots. Dins d'una missió, les tasques de modelatge i interacció amb l'entorn defineix les capacitats d'operació dels robots dins d'una missió. Mitjançant la integració d'un grup heterogeni de robots seves diverses capacitats són complementades per a l'assoliment un objectiu comú. Per tal de caracteritzar aquesta proposta, els mecanismes presentats en aquest document s'avaluaran en detall mitjançant d'una sèrie experiments en els quals un grup de robots heterogenis executen una missió col·laborativa per aconseguir un objectiu comú. Finalment, els resultats seran discutits a manera de conclusions donant lloc a l'establiment d'un treball futur.
Munera Sánchez, E. (2017). MISSION-ORIENTED HETEROGENEOUS ROBOT COOPERATION BASED ON SMART RESOURCES EXECUTION [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/88404
TESIS

Este proyecto se centra en el estudio de la empresa “Pacifico Seguros” que opera principalmente en el sector de seguros. El desarrollo del trabajo está enfocado en el subproceso de “Emisión” que es uno de los procesos importantes de la empresa porque acá se inicia todo el proceso de las emisiones de los clientes para las afiliaciones de sus trabajadores. La oportunidad de mejora se presenta en la mayoría de actividades del subproceso de “Emisión” que suelen ser repetitivas y operativas; y en su mayoría son ejecutadas de forma manual. Por tal motivo, el proyecto contempla una solución para la optimización del sub proceso que permitirá reducir el tiempo de atención como principal objetivo. La propuesta consta de un software RPA que permita a los usuarios crear “bots” que pueden aprender, imitar y luego ejecutar procesos del negocio basados en reglas para la gestión administrativa que ayudará a la descarga y respuestas de los correos, también contará con un motor de asignación para la distribución equitativa y atención de las solicitudes de afiliación. El propósito del trabajo es llevar a cabo la gestión del desarrollo del software y la arquitectura empresarial dentro del marco de trabajo de TOGAF y PMBOK sobre el objeto estudio para la evaluación de los procesos y para una correcta arquitectura de software lo cual se podrá desarrollar en fases y actividades para formular el plan estratégico y el control de su cumplimiento.
This project focuses on the study of the company “Pacificos Seguros” that operates mainly in the insurance sector. The development of the work focuses on the sub-process “Emisión” which is one of the important processes of the company because here begins the process of customer emissions for the affiliations of its workers. The opportunity for improvement occurs in most of the activities of the sub-process “Emisión”, that tend to be repetitive and operational and they are mostly run manually. Therefore, the project contemplates a solution for the optimization of the sub-process that will reduce the attention time as main objective. The proposal consists of an RPA software that allows users to create "bots" that can learn, imitate and then execute business processes based on rules for the administrative management that help download and reply to emails also there will be a queue engine for the distribution and attention of the affiliation requests. The purpose of the work is to carry out the management of the software development and business architecture in the TOGAF and PMBOK framework on the study object for the evaluation of processes and a correct software architecture which can will be developed in phases and activities to formulate a strategic plan and control its compliance.
Tesis

Les robots sous-marins peuvent aujourd'hui évoluer dans des environnements complexes difficilement accessibles à l'Homme pour des raisons de coût ou de sécurité. Ils peuvent donc intervenir dans une grande variété de missions en environnement aquatique. Or, la complexité de ces milieux impose de doter le vecteur robotique d'une autonomie opérationnelle suffisante afin qu'il puisse mener sa mission à bien tout en préservant son intégrité. Cela nécessite de développer des lois de commande répondant aux spécificités de l'application. Ces lois de commande se basent sur des connaissances provenant de différentes disciplines scientifiques ce qui souligne l'interdisciplinarité inhérente à la robotique. Une fois la loi de commande développée, il faut implémenter le contrôleur sur le robot sous forme de logiciel de contrôle basé sur une architecture logicielle temps-réel.Or la conception actuelle des lois de commande, sous forme de blocs "monolithiques", rend difficile l'évolution d'une loi de commande d'une application à l'autre, l'intégration de connaissances provenant d'autres disciplines scientifiques que ne maitrisent pas forcément les automaticiens et pénalisent son implémentation sur des architectures logicielles qui nécessitent la modularité. Pour résoudre ces problèmes nous cherchons à proprement séparer les différentes connaissances afin que chacune soit aisément manipulable, son rôle clair et que les relations établies entre les différentes connaissances soient explicites. Cela permettra en outre une projection plus efficace sur l'architecture logicielle. Nous proposons donc un nouveau formalisme de description des lois de commande selon une composition modulaire d'entités de base appelées Atomes et qui encapsulent les différents éléments de connaissance. Nous nous intéressons également à l'établissement d'une meilleure synergie entre les aspects automatique et génie logiciel qui se construit autour de préoccupations communes telles que les contraintes temporelles et la stabilité. Pour cela, nous enrichissons nos Atomes de contraintes chargées de véhiculer les informations relatives à ces aspects temporels. Nous proposons également une méthodologie basée sur notre formalisme afin de guider l'implémentation de nos stratégies de commande sur un Middleware temps-réel, dans notre cas le Middleware ContrACT développé au LIRMM.Nous illustrons notre approche par diverses fonctionnalités devant être mises en oeuvre lors de missions d'exploration de l'environnement aquatique et notamment pour l'évitement de parois lors de l'exploration d'un aquifère karstique
Underwater robots can nowadays operate in complex environments in a broad scope of missions where the use of human divers is difficult for cost or safety reasons. However the complexity of aquatic environments requires to give the robotic vector an autonomy sufficient to perform its mission while preserving its integrity. This requires to design control laws according to application requirements. They are built on knowledge from several scientific fields, underlining the interdisciplinarity inherent to robotics. Once the control law designed, it must be implemented as a control Software working on a real-time Software architecture.Nonetheless the current conception of control laws, as "monolithic" blocks, makes difficult the adaptation of a control from an application to another and the integration of knowledge from various scientific fields which are often not fully understood by control engineers. It also penalizes the implementation of control on Software architectures, at least its modularity and evolution. To solve those problems we seek a proper separation of knowledge so that each knowledge item can be easily used, its role precisely defined and we want to reify the interactions between them. Moreover this will allow us a more efficient projection on the Software architecture. We thus propose a new formalism for control laws description as a modular composition of basic entities named Atoms used to encapsulate the knowledge items.We also aim at building a better synergy between control and software engineering based on shared concerns such as temporal constraints and stability. Hence we extend the definition of our Atoms with constraints carrying information related to their temporal behaviour. We propose as well a methodology relying on our formalism to guide the implementation of control on a real-time Middleware. We will focus on the ContrACT Middleware developed at LIRMM.Finally we illustrate our approach on several robotic functionalities that can be used during aquatic environments exploration and especially for wall avoidance during the exploration of a karst aquifer

The dissemination of new innovative technology requires feasibility and simplicity. The problem with marker-based augmented reality is similar to glove-based hand gesture recognition: they both require an additional component to function. This thesis investigates the possibility of combining markerless augmented reality together with appearance-based hand gesture recognition by implementing a game with good user experience. The methods employed in this research consist of a game implementation and a pre-study meant for measuring interactive accuracy and precision, and for deciding upon which gestures should be utilized in the game. A test environment was realized in Unity using ARKit and Manomotion SDK. Similarly, the implementation of the game used the same development tools. However, Blender was used for creating the 3D models. The results from 15 testers showed that the pinching gesture was the most favorable one. The game was evaluated with a System Usability Scale (SUS) and received a score of 70.77 among 12 game testers, which indicates that the augmented reality game, which interaction method is solely based on bare-hands, can be quite enjoyable.

Self-adaptive systems have limited time to adjust their configurations whenever their adaptation goals, i.e., quality requirements, are violated due to some runtime uncertainties. Within the available time, they need to analyze their adaptation space, i.e., a set of configurations, to find the best adaptation option, i.e., configuration, that can achieve their adaptation goals. Existing formal analysis approaches find the best adaptation option by analyzing the entire adaptation space. However, exhaustive analysis requires time and resources and is therefore only efficient when the adaptation space is small. The size of the adaptation space is often in hundreds or thousands, which makes formal analysis approaches inefficient in large-scale self-adaptive systems. In this thesis, we tackle this problem by presenting an online learning approach that enables formal analysis approaches to analyze large adaptation spaces efficiently. The approach integrates with the standard feedback loop and reduces the adaptation space to a subset of adaptation options that are relevant to the current runtime uncertainties. The subset is then analyzed by the formal analysis approaches, which allows them to complete the analysis faster and efficiently within the available time. We evaluate our approach on two different instances of an Internet of Things application. The evaluation shows that our approach dramatically reduces the adaptation space and analysis time without compromising the adaptation goals.

Research Doctorate - Doctor of Philosophy (PhD)
This thesis identifies the restrictions that resource-constrained robotic platforms experience in relation to their software architecture and computer vision. It proposes and evaluates a number of techniques to support more effective implementations. A review of robotic software shows that they often implement a message passing framework as their software architecture. This supports maintainability and modularity. However, the loose coupling afforded by message passing systems has a computational cost on performance and inhibits the ease of accessing multiple data sources. Resource-constrained robotic systems do not have the computational power available to support real-time application of a message passing framework. To resolve this, two compile time message routing techniques, Typemap Routing and Link Time Route Generation (LTRG), are presented and evaluated. They are compared against a hashmap and treemap in both single and multithreaded environments and were shown to improve predictability and latency. A technique called co-messaging is identified to reduce complexity in message passing systems through easier data access. Using Typemap Routing and co-messaging, NUClear, a software architecture framework was developed. NUClear's execution time was compared to Robot Operating System (ROS) and through an implementation of NUClear in the NUbots' codebase. NUClear's interface complexity and memory usage were evaluated, showing improvements. The implementation of co-messaging and Typemap Routing supported the development of an Adaptive Lookup Table. It was developed to resolve the issues that resource-constrained robotic systems experience when classifying colours in various lighting conditions. It was assessed against and outperformed a static lookup table. A Visual Mesh was developed to improve computer vision and reduce the computational cost of convolutional neural networks on resource-constrained robots. It was evaluated against a hexagonal mesh and other convolutional neural networks and resulted in improved accuracy and execution performance. Its performance over distance exceeded more complex networks. The techniques developed and the evaluation presented in this thesis are important for the fields of software architecture and computer vision. They offer solutions to improve performance for both resource constrained robotic systems as well as less constrained systems.

M.Sc. (Computer Science)
General machine intelligence represents the principal ambition of artificial intelligence research: creating machines that readily adapt to their environment. Machine learning represents the driving force of adaptation in artificial intelligence. However, two pertinent dilemmas emerge from research into machine learning. Firstly, how do intelligent agents learn effectively in real-world environments, in which randomness, perceptual aliasing and dynamics complicate learning algorithms? Secondly, how can intelligent agents exchange knowledge and learn from one another without introducing mathematical anomalies that might impede on the effectiveness of the applied learning algorithms? In a robotic search and rescue scenario, for example, the control system of each robot must learn from its surroundings in a fast-changing and unpredictable environment while at the same time sharing its learned information with others. In well-understood problems, an intelligent agent that is capable of solving task-specific problems will suffice. The challenge behind complex environments comes from fact that agents must solve arbitrary problems (Kaelbling et al. 1996; Ryan 2008). General problem-solving abilities are hence necessary for intelligent agents in complex environments, such as robotic applications. Although specialized machine learning techniques and cognitive hierarchical planning and learning may be a suitable solution for general problem-solving, such techniques have not been extensively explored in the context of cooperative multi-agent learning. In particular, to the knowledge of the author, no cognitive architecture has been designed which can support knowledge-sharing or self-organisation in cooperative multi-agent learning systems. It is therefore social learning in real-world applications that forms the basis of the research presented in this dissertation. This research aims to develop a distributed cognitive architecture for cooperative multi-agent learning in complex environments. The proposed Multi-agent Learning through Distributed Adaptive Contextualization Distributed Cognitive Architecture for Multi-agent Learning (MALDAC) Architecture comprises a self-organising multi-agent system to address the communication constraints that the physical hardware imposes on the system. The individual agents of the system implement their own cognitive learning architecture. The proposed Context-based Adaptive Empathy-deliberation Agent (CAEDA) Architecture investigates the applicability of emotion, ‘consciousness’, embodiment and sociability in cognitive architecture design. Cloud computing is proposed as a method of service delivery for the learning system, in which the MALDAC Architecture governs multiple CAEDA-based agents. An implementation of the proposed architecture is applied to a simulated multi-robot system to best emulate real-world complexities. Analyses indicate favourable results for the cooperative learning capabilities of the proposed MALDAC and CAEDA architectures.