This thesis presents a novel design paradigm of mobile robots: the Hybrid Mobile Robot system. It consists of a combination of parallel and serially connected links resulting in a hybrid mechanism that includes a mobile robot platform for locomotion and a manipulator arm for manipulation, both interchangeable functionally. All state-of-the-art mobile robots have a separate manipulator arm module attached on top of the mobile platform. The platform provides mobility and the arm provides manipulation. Unlike them, the new design has the ability to interchangeably provide locomotion and manipulation capability, both simultaneously. This was accomplished by integrating the locomotion platform and the manipulator arm as one entity rather than two separate and attached modules. The manipulator arm can be used as part of the locomotion platform and vice versa. This paradigm significantly enhances functionality. The new mechanical design was analyzed with a virtual prototype that was developed with MSC Adams Software. Simulations were used to study the robot’s enhanced mobility through animations of challenging tasks. Moreover, the simulations were used to select nominal robot parameters that would maximize the arm’s payload capacity, and provide for locomotion over unstructured terrains and obstacles, such as stairs, ditches and ramps. The hybrid mobile robot also includes a new control architecture based on embedded on-board wireless communication network between the robot’s links and modules such as the actuators and sensors. This results in a modular control architecture since no cable connections are used between the actuators and sensors in each of the robot links. This approach increases the functionality of the mobile robot also by providing continuous rotation of each link constituting the robot. The hybrid mobile robot’s novel locomotion and manipulation capabilities were successfully experimented using a complete physical prototype. The experiments provided test results that support the hypothesis on the qualitative and quantitative performance of the mobile robot in terms of its superior mobility, manipulation, dexterity, and ability to perform very challenging tasks. The robot was tested on an obstacle course consisting of various test rigs including man–made and natural obstructions that represent the natural environments the robot is expected to operate on.

This thesis presents a novel design paradigm of mobile robots: the Hybrid Mobile Robot system. It consists of a combination of parallel and serially connected links resulting in a hybrid mechanism that includes a mobile robot platform for locomotion and a manipulator arm for manipulation, both interchangeable functionally. All state-of-the-art mobile robots have a separate manipulator arm module attached on top of the mobile platform. The platform provides mobility and the arm provides manipulation. Unlike them, the new design has the ability to interchangeably provide locomotion and manipulation capability, both simultaneously. This was accomplished by integrating the locomotion platform and the manipulator arm as one entity rather than two separate and attached modules. The manipulator arm can be used as part of the locomotion platform and vice versa. This paradigm significantly enhances functionality. The new mechanical design was analyzed with a virtual prototype that was developed with MSC Adams Software. Simulations were used to study the robot’s enhanced mobility through animations of challenging tasks. Moreover, the simulations were used to select nominal robot parameters that would maximize the arm’s payload capacity, and provide for locomotion over unstructured terrains and obstacles, such as stairs, ditches and ramps. The hybrid mobile robot also includes a new control architecture based on embedded on-board wireless communication network between the robot’s links and modules such as the actuators and sensors. This results in a modular control architecture since no cable connections are used between the actuators and sensors in each of the robot links. This approach increases the functionality of the mobile robot also by providing continuous rotation of each link constituting the robot. The hybrid mobile robot’s novel locomotion and manipulation capabilities were successfully experimented using a complete physical prototype. The experiments provided test results that support the hypothesis on the qualitative and quantitative performance of the mobile robot in terms of its superior mobility, manipulation, dexterity, and ability to perform very challenging tasks. The robot was tested on an obstacle course consisting of various test rigs including man–made and natural obstructions that represent the natural environments the robot is expected to operate on.

The field of autonomous vehicle navigation and localization is a highly active research topic. The aim of this thesis is to evaluate the feasibility to use outdoor visual navigation in a semi-structured environment. The goal is to develop a visual navigation system for an autonomous golf ball collection vehicle operating on driving ranges. The image feature extractors SIFT and PCA-SIFT was evaluated on an image database consisting of images acquired from 19 outdoor locations over a period of several weeks to allow different environmental conditions. The results from these tests show that SIFT-type feature extractors are able to find and match image features with high accuracy. The results also show that this can be improved further by a combination of a lower nearest neighbour threshold and an outlier rejection method to allow more matches and a higher ratio of correct matches. Outliers were found and rejected by fitting the data to a homography model with the RANSAC robust estimator algorithm. A simulator was developed to evaluate the suggested system with respect to pixel noise from illumination changes, weather and feature position accuracy as well as the distance to features, path shapes and the visual servoing target image (milestone) interval. The system was evaluated on a total of 3 paths, 40 test combinations and 137km driven. The results show that with the relatively simple visual servoing navigation system it is possible to use mono-vision as a sole sensor and navigate semi-structured outdoor environments such as driving ranges.

In this thesis, I study the computational advantages of the allocentric represen- tation as compared to the egocentric representation for autonomous local navigation. Whereas in the allocentric framework, all variables of interest are represented with respect to a coordinate frame attached to an object in the scene, in the egocentric one, they are always represented with respect to the robot frame at each time step. In contrast with well-known results in the Simultaneous Localization and Mapping literature, I show that the amounts of nonlinearity of these two representations, where poses are elements of Lie-group manifolds, do not affect the accuracy of Gaussian- based filtering methods for perception at both the feature level and the object level. Furthermore, although these two representations are equivalent at the object level, the allocentric filtering framework is better than the egocentric one at the feature level due to its advantages in the marginalization process. Moreover, I show that the object- centric perspective, inspired by the allocentric representation, enables novel linear- time filtering algorithms, which significantly outperform state-of-the-art feature-based filtering methods with a small trade-off in accuracy due to a low-rank approximation. Finally, I show that the allocentric representation is also better than the egocentric representation in Model Predictive Control for local trajectory planning and obstacle avoidance tasks.

In autonomous navigation of mobile robots the dynamic environment is a source of problems. Because it is not possible to model all the possible conditions, the key point in the robot control is to design a system that is adaptable to different conditions and robust in dynamic environments. This study presents a reactive control system for a Khepera robot with the ability to navigate in a dynamic environment for reaching goal objects. The main motivation of this research is to design a robot control, which is robust to sensor errors and sudden changes and adaptable to different environments and conditions. Behavior based approach is used with taking the advantage of fuzzy reasoning in design. Experiments are made on Webots, which is a simulation environment for Khepera robot.

In agent based systems, especially in autonomous mobile robots, modelling the environment and its changes is a source of problems. It is not always possible to effectively model the uncertainity and the dynamic changes in complex, real-world domains. Control systems must be robust to changes and must be able to handle these uncertainties to overcome this problem. In this study, a reactive behaviour based agent control system is modelled and implemented. The control system is tested in a navigation task using an environment, which has randomly placed obstacles and a goal position to simulate an environment similar to an autonomous robot&rsquo / s indoor environment. Then the control system was extended to control an agent in a multi-agent environment. The main motivation of this study is to design a control system which is robust to errors and easy to modify. Behaviour based approach with the advantages of fuzzy reasoning systems is used in the system.

abstract: Advancements in computer vision and machine learning have added a new dimension to remote sensing applications with the aid of imagery analysis techniques. Applications such as autonomous navigation and terrain classification which make use of image classification techniques are challenging problems and research is still being carried out to find better solutions. In this thesis, a novel method is proposed which uses image registration techniques to provide better image classification. This method reduces the error rate of classification by performing image registration of the images with the previously obtained images before performing classification. The motivation behind this is the fact that images that are obtained in the same region which need to be classified will not differ significantly in characteristics. Hence, registration will provide an image that matches closer to the previously obtained image, thus providing better classification. To illustrate that the proposed method works, naïve Bayes and iterative closest point (ICP) algorithms are used for the image classification and registration stages respectively. This implementation was tested extensively in simulation using synthetic images and using a real life data set called the Defense Advanced Research Project Agency (DARPA) Learning Applied to Ground Robots (LAGR) dataset. The results show that the ICP algorithm does help in better classification with Naïve Bayes by reducing the error rate by an average of about 10% in the synthetic data and by about 7% on the actual datasets used. / Dissertation/Thesis / M.S. Electrical Engineering 2011

Veículos autônomos terrestres são um tipo de veículo motorizado guiado de forma autônoma por um sistema computacional, sem necessidade de intervenção humana. Esta pesquisa analisa a Tomada de Decisão autônoma necessária para este tipo de veículo, assim como os processos de navegação e as estruturas de dados que a suportam. Sobre esta análise propõe-se uma arquitetura de Tomada de Decisão para navegação autônoma baseada em dois ideias principais: a sistematização das fontes de dados com as quais o veículo deve interagir em sua navegação em uma hierarquia de prioridades de tratamento e o desenvolvimento de uma estrutura de dados chamada de Grade de Prioridades que permite definir para qual região do ambiente o veículo deve se locomover. Para suportar a arquitetura, formaliza-se um conjunto de processos recomendados de navegação robótica (percepção, posicionamento, planejamento e estratégia de movimento), organizados para suportar a integração com recursos de outras pesquisas do segmento da Robótica Móvel. Para validação da arquitetura, desenvolve-se um sistema de navegação robótica, integrada à uma plataforma robótica real, e realiza-se um conjunto de testes para simular o comportamento da arquitetura em situações de navegação específicas. / Autonomous ground vehicles are a type of motor vehicle autonomously guided by a computer system without human intervention. This research analyzes the autonomous decision making required for this type of vehicle as well as the processes of navigation and data structures that support it. On this analysis, is proposed an architecture of decision making for autonomous navigation based on two fundamental ideas: the systematization of data sources with which a vehicle must interact in their navigation in a hierarchy of priorities and the development of a special data structure called Grid Priorities that defines which region of the environment the vehicle must be moved. Supporting the architecture, a set of processes for robotic navigation (perception, positioning, planning and motion strategy) are proposed and organized to support integration with other research resources in the segment of Mobile Robotics. For validate the architecture, is developed a system of robotic navigation, integrated with a real robotic platform, and is performed a set of tests to simulate the behavior of the architecture in specific situations of navigation.

Orientadores: Janito Vaqueiro Ferreira, Arthur de Miranda Neto / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-26T02:40:48Z (GMT). No. of bitstreams: 1 RodriguezRuiz_MariaFernanda_M.pdf: 8751418 bytes, checksum: adc682edc93b0bb96046bdc3bde2c72e (MD5) Previous issue date: 2014 / Resumo: Os veiculos autônomos são uma realidade, mas seu desenvolvimento requer a altos custos. No entanto, a fim de fortalecer os avanços na robótica móvel, há um grande esforço no desenvolvimento de aplicações de baixo custo orientadas aos veículos autônomos, sendo o estudo de métodos de localização uma das áreas de maior interesse, a fim de uma navegação segura. Esta dissertação de mestrado propõe um método de localização híbrida, composto por sensores proprioceptivos e exteroceptivos, e informações adicionais como mapas digitais e reconhecimento de objetos chave, para melhorar a estimação da posição de um veículo terrestre. O sistema está baseado nas informações de posicionamento obtidas por um GPS de baixo custo e também em informações obtidas pelo método de localização referenciada, baseada em um conjunto de dados geográficos disponíveis em uma base de dados que compõe o mapa digital. A técnica de fusão de dados selecionada é baseada no filtro de Kalman estendido (EKF), que combina as informações de diferentes sistemas a fim de aumentar a robustez do sistema de localização. O desempenho do método de localização proposto é verificado através de uma plataforma de localização composta por ferramentas como um servidor de mapas baseado em OpenStreetMap e uma outra de simulação composta por ferramentas como ROS e Gazebo, que inclui um veículo terrestre com sensores embarcados, um mapa digital e objetos chave. A arquitetura selecionada é de fácil adequação para a realização de testes, já que permite simular objetos e sensores bem próximos à realidade / Abstract: The autonomous vehicles are a reality, but its development is equivalent to high costs. However, in order to strengthen the advances in mobile robotics, there is a large effort in developing low cost oriented applications to autonomous vehicles, with the study of methods of finding the area of most interest in relation to the fact achieve a navigation secure. This dissertation proposes a method of hybrid location consists of proprioceptive and exteroceptive sensors, and additional information such as digital maps and recognition of key objects, to improve the estimation of a land vehicle position. The system is based on the positioning information via GPS low cost, and information obtained by the method referenced location, based on a set of available spatial data in a database comprising the digital map. The fusion technique selected data is based on the extended Kalman filter (EKF) that combines the information of different systems in order to increase the robustness of the location system. The performance of the proposed method of location is verified through a location platform consists of a server tools like maps based on OpenStreetMap and another comprising simulation tools such as ROS and Gazebo, which includes a ground vehicle with embedded sensors, a map digital and key objects. The selected architecture is easily suitability for testing, since it allows you to simulate objects and sensors very close to reality / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestra em Engenharia Mecânica

Orientador: José Raimundo de Oliveira / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-23T02:36:38Z (GMT). No. of bitstreams: 1 Santos_MiltonFelipeSouza_M.pdf: 3219713 bytes, checksum: c54c1e303b8b3f5087f884140423e2a6 (MD5) Previous issue date: 2013 / Resumo: Este trabalho analisa as possibilidades de arquiteturas de hardware buscando encontrar uma plataforma de desenvolvimento escalável e possível de se adaptar a diferentes estruturas mecânicas de veículos móveis. Esta plataforma deve ainda ter sensores suficientes para permitir comportamentos autônomos ao veículo. Para isto, de forma a entender melhor o conceito de autonomia, primeiramente foram analisadas as capacidades mentais humanas. Das capacidades estudadas foram escolhidas as capacidades de sensação, percepção, orientação e cognição como capacidades relacionadas ao hardware quando implantadas em veículos móveis artificialmente. Em seguida cada uma destas quatro capacidades mentais foi analisada a fundo buscando métodos e sistemas para solucionar estes problemas. Foram analisadas também as possibilidades de topologias em rede de forma a conectar os módulos individuais e propostos critérios de escolha dos módulos e topologias do sistema. Após todas estas análises exaustivas, onde somente as partes mais importantes foram incluídas neste trabalho, foi escolhida a topologia de barramento serial com arbitração em hardware para permitir modularidade, escalabilidade e confiabilidade. Como barramento elétrico foi escolhido o barramento CAN (Controller Area Network) que juntamente com diversos módulos especificados neste trabalho resultam na plataforma que foi chamada "Sistema Autônomo Completo". Com base neste sistema, foram propostos como resultados duas simplificações: uma baseada em veículos que operam no solo (bidimensional) e outra para veículos que operam em ambientes tridimensionais como o ar ou a água. Buscando um enfoque educacional foi proposta uma sequência de implantação do sistema autônomo completo e alguns estudos de casos estão apresentados. Com este trabalho foi possível estruturar e criar uma sequência de desenvolvimento de um veículo móvel robótico em fases que podem ser facilmente seguidas por escolas ou empresas / Abstract: This work analyzes hardware architectures of embedded systems, searching for a development platform of mobile robots. This platform must be scalable and easily adaptable to several types of mechanical designs of mobile vehicles. And it must have enough sensors in order to reach autonomous performance. For better understanding concepts of autonomy, the human brain capacities were studied. From the studied capacities, sensation, perception, representation and orientation were chosen. These four capacities were chosen as related with hardware implementations if artificially designed for mobile vehicles. Network topologies for connecting modules of independent methods for the artificial brain capabilities mentioned were also analyzed. For choosing the best proposal, some criteria were defined for the modules and system topologies. After this analysis, where only the most important parts were included, a topology was chosen. The chosen topology is the serial bus with hardware arbitration. The chosen electrical bus was the CANbus, which together with the other modules specified in this work resulted in the platform called "Full Autonomous Vehicle". Based on this system platform, simplifications were proposed: one focused on vehicles with two-dimensional movements, and other focused on vehicles with three-dimensional movements. Searching for an educational point-of-view, an implementation sequence was proposed for the full autonomous vehicle and some cases were studied. With this work was possible to organize and create a development sequence of a robotic mobile vehicle divided by phases. These phases can be easily followed by schools and companies / Mestrado / Automação / Mestre em Engenharia Elétrica