Autonomous underwater vehicle control using reinforcement learning policy search methods

Autonomous underwater vehicles (AUV) represent a challenging control problem with complex, noisy, dynamics. Nowadays, not only the continuous scientific advances in underwater robotics but the increasing number of subsea missions and its complexity ask for an automatization of submarine processes. This paper proposes a high-level control system for solving the action selection problem of an autonomous robot. The system is characterized by the use of reinforcement learning direct policy search methods (RLDPS) for learning the internal state/action mapping of some behaviors. We demonstrate its feasibility with simulated experiments using the model of our underwater robot URIS in a target following task

Plan de empresa "Dorcé"

Dorcé es una idea de negocio que responde de manera directa a la necesidad de vestir de las mujeres a través de su innovador sistema operativo “Diseño y confección a tu manera” el cual tiene como objetivo revolucionar la tradicional forma de adquirir prendas de vestir femeninas en la ciudad de Villavicencio. Desde su creación y ejecución, se ha observado una gran aceptación de dicho sistema operativo por parte de los clientes antiguos generando muy buenos resultados económicos partiendo de sus actividades principales, el diseño, la confección y venta de prendas de vestir femeninas. Es por esto que el desarrollo de este trabajo de grado va dirigido a organizar de una manera más administrativa los diferentes procesos que se llevan a cabo dentro de la organización a fin de conseguir incrementar la producción y llegar así a un grupo más amplio de clientes con base en sus necesidades y a las actuales tendencias de la moda. Esta labor tiene como fin, responder y satisfacer a la mujer villavicense en términos de moda, servicio y confección, agregándole un valor adicional en cuanto a menores tiempos de entrega, altos estándares de calidad y exclusividad. Todo lo anterior, en un solo lugar. El sistema operativo (factor diferenciador) de esta idea de negocio consiste en ofrecer a los clientes, diferentes modelos de las partes (cuello, manga, contorno, falda y panty) que componen las prendas que fabrica Dorcé (Blusa, vestido y traje de baño). Con el fin de que el cliente pueda armar la prenda a su gusto teniendo el plus de ir “a la medida”. Dicha prenda de carácter exclusivo, es confeccionada y entregada al cabo de un máximo de 3 días dependiendo de la complejidad que tenga para ser elaborada, garantizándole al cliente no solo la originalidad del diseño de su prenda sino también fabricarla en menor tiempo que la competencia. Es un producto y un servicio innovador que va dirigido a mujeres de quince a setenta años de edad de la ciudad de Villavicencio, mujeres de media y alta sociedad que se caracterizan por su gusto por la moda y la calidad. Una vez se satisfacen las necesidades de nuestros clientes se da lugar a nuevas ventas dentro del grupo social de cada una de ellas, una vez nos refieran de manera positiva con sus amigas y conocidas. Este tipo de publicidad (voz a voz) es la que ha llevado a Dorcé a crecer a través de los años sin perder la exclusividad de su mercado; sin embargo, en la actualidad, la idea de negocio carece de estructura empresarial adecuada, de estrategias que garanticen su crecimiento y sostenibilidad, también carece de recursos financieros inmediatos que faciliten la ampliación y adecuación de la planta de producción y del almacén. Es por ello que las actuales empresarias de esta idea de negocio tienen como objetivo, mediante la elaboración de este plan de empresa, desarrollar, organizar y mejorar las estrategias, acciones, objetivos, alianzas, cambios, inversiones, que se tienen y necesitan para el mejoramiento interno y externo de la compañía. Para el desarrollo de dicho plan, Dorcé cuenta con Doris Blanco, una diseñadora de modas de 25 años de experiencia en el sector textil y confección en la ciudad de Villavicencio, quien está a cargo del área de producción de la empresa, dos estudiantes de Administración de Negocios Internacionales, Fernanda Montes y Juliana Villamizar, quienes están a cargo de la parte administrativa, comercial y estratégica de la empresa. De igual forma cuenta con el conocimiento del mercado textil y de confección, así mismo amplias relaciones que se han venido fortaleciendo a través del tiempo con los diferentes proveedores de materia prima en la ciudad, maquinaria y materia prima inicial, cerca de sesenta mujeres que se han vinculado a la empresa como clientes fieles a la idea de negocio promocionándolo de manera positiva día a día, así como también el interés, empuje y compromiso por parte de las involucradas en desarrollar, ejecutar y sostener esta idea de negocio como proyecto de vida.

Modeling camera orientation and 3D structure from a sequence of images taken by a perambulating commercial video camera

In this paper we report the degree of reliability of image sequences taken by off-the-shelf TV cameras for modeling camera rotation and reconstructing 3D structure using computer vision techniques. This is done in spite of the fact that computer vision systems usually use imaging devices that are specifically designed for the human vision. Our scenario consists of a static scene and a mobile camera moving through the scene. The scene is any long axial building dominated by features along the three principal orientations and with at least one wall containing prominent repetitive planar features such as doors, windows bricks etc. The camera is an ordinary commercial camcorder moving along the axial axis of the scene and is allowed to rotate freely within the range +/- 10 degrees in all directions. This makes it possible that the camera be held by a walking unprofessional cameraman with normal gait, or to be mounted on a mobile robot. The system has been tested successfully on sequence of images of a variety of structured, but fairly cluttered scenes taken by different walking cameramen. The potential application areas of the system include medicine, robotics and photogrammetry.

Formação da habilidade de interpretar gráficos cartesianos em licenciandos em química segundo a teoria de P. Ya. Galperin

It has been remarkable among the Science Teaching debates the necessity that students do not learn only theories, laws and concepts, but also develop skills which allows them to act towards a critical citizenship. Therefore, some of the skills for the natural sciences learning must be taught consciously, intentionally and in a planned way, as component of a basic competence. Studies of the last twenty years have shown that students and teachers have plenty of difficulties about skills development and, among several, the skill of interpreting Cartesian graphics, essential for the comprehension of Natural Science. In that sense, the development of that type of professional knowledge during the initial education of future Chemistry teachers has become strategic, not only because they need to know how to use it, but also because they need to know how to teach it. This research has as its general objective the organization, development and study of a process of formation of the skill of interpreting Cartesian graphics as part of the teachers professional knowledge. It has been accomplished through a formative experience with six undergraduate students of the Teaching Degree Course of Chemistry of Universidade Federal do Rio Grande do Norte (UFRN Federal University of Rio Grande do Norte), in Brazil. In order to develop that skill, we have used as reference P. Ya. Galperin s Theory of the Stepwise Formation of Mental Actions and Concepts and its following qualitative indicators: action form, degree of generalization, degree of consciousness, degree of independence and degree of solidness. The research, in a qualitative approach, has prioritized as instruments of data collecting the registering of the activities of the undergraduate students, the observation, the questionnaire and the diagnosis tests. At the first moment, a teaching framework has been planned for the development of the skill of interpreting Cartesian graphics based on the presupposed conceptions and steps of Galperin s Theory. At the second moment, the referred framework has been applied and the process of the skill formation has been studied. The results have shown the possibility of develop the skill conscious about the invariant operation system, with a high degree of generalization and internalized the operational invariant in the mental plane. The students have attested the contributions at that type of formative experience. The research reveals the importance of going deeper about the teaching comprehension of the individualities tied to the process of internalization, according to Galperin s Theory, when the update of abilities as part of the teaching professional knowledge is the issue

Sistema especialista para detecção de falhas em comandos elétricos

This work consists of the creation of a Specialist System which utilizes production rules to detect inadequacies in the command circuits of an operation system and commands of electric engines known as Direct Start. Jointly, three other modules are developed: one for the simulation of the commands diagram, one for the simulation of faults and another one for the correction of defects in the diagram, with the objective of making it possible to train the professionals aiming a better qualification for the operation and maintenance. The development is carried through in such a way that the structure of the task allows the extending of the system and a succeeding promotion of other bigger and more complex typical systems. The computational environment LabView is employed to enable the system

Angiotensin II-derived reactive oxygen species underpinning the processing of the cardiovascular reflexes in the medulla oblongata

The brainstem is a major site in the central nervous system involved in the processing of the cardiovascular reflexes such as the baroreflex and the peripheral chemoreflex. The nucleus tractus solitarius and the rostral ventrolateral medulla are 2 important brainstem nuclei, and they play pivotal roles in autonomic cardiovascular regulation. Angiotensin II is one of the neurotransmitters involved in the processing of the cardiovascular reflexes within the brainstem. It is well-known that one of the mechanisms by which angiotensin II exerts its effect is via the activation of pathways that generate reactive oxygen species (ROS). In the central nervous system, ROS are reported to be involved in several pathological diseases such as hypertension, heart failure and sleep apnea. However, little is known about the role of ROS in the processing of the cardiovascular reflexes within the brainstem. The present review mainly discussed some recent findings documenting a role for ROS in the processing of the baroreflex and the peripheral chemoreflex in the brainstem.

Carregamento ótimo de veículos elétricos considerando as restrições das redes de distribuição de média tensão

Pós-graduação em Engenharia Elétrica - FEIS

Navegação de robôs em ambientes internos usando slam

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Uso de árvore de decisão para avaliação da segurança estática em tempo real de sistemas elétricos de potência

As técnicas utilizadas para avaliação da segurança estática em sistemas elétricos de potência dependem da execução de grande número de casos de fluxo de carga para diversas topologias e condições operacionais do sistema. Em ambientes de operação de tempo real, esta prática é de difícil realização, principalmente em sistemas de grande porte onde a execução de todos os casos de fluxo de carga que são necessários, exige elevado tempo e esforço computacional mesmo para os recursos atuais disponíveis. Técnicas de mineração de dados como árvore de decisão estão sendo utilizadas nos últimos anos e tem alcançado bons resultados nas aplicações de avaliação da segurança estática e dinâmica de sistemas elétricos de potência. Este trabalho apresenta uma metodologia para avaliação da segurança estática em tempo real de sistemas elétricos de potência utilizando árvore de decisão, onde a partir de simulações off-line de fluxo de carga, executadas via software Anarede (CEPEL), foi gerada uma extensa base de dados rotulada relacionada ao estado do sistema, para diversas condições operacionais. Esta base de dados foi utilizada para indução das árvores de decisão, fornecendo um modelo de predição rápida e precisa que classifica o estado do sistema (seguro ou inseguro) para aplicação em tempo real. Esta metodologia reduz o uso de computadores no ambiente on-line, uma vez que o processamento das árvores de decisão exigem apenas a verificação de algumas instruções lógicas do tipo if-then, de um número reduzido de testes numéricos nos nós binários para definição do valor do atributo que satisfaz as regras, pois estes testes são realizados em quantidade igual ao número de níveis hierárquicos da árvore de decisão, o que normalmente é reduzido. Com este processamento computacional simples, a tarefa de avaliação da segurança estática poderá ser executada em uma fração do tempo necessário para a realização pelos métodos tradicionais mais rápidos. Para validação da metodologia, foi realizado um estudo de caso baseado em um sistema elétrico real, onde para cada contingência classificada como inseguro, uma ação de controle corretivo é executada, a partir da informação da árvore de decisão sobre o atributo crítico que mais afeta a segurança. Os resultados mostraram ser a metodologia uma importante ferramenta para avaliação da segurança estática em tempo real para uso em um centro de operação do sistema.

Comparing support vector machines and artificial neural networks in the recognition of steering angle for driving of mobile robots through paths in plantations

The use of mobile robots turns out to be interesting in activities where the action of human specialist is difficult or dangerous. Mobile robots are often used for the exploration in areas of difficult access, such as rescue operations and space missions, to avoid human experts exposition to risky situations. Mobile robots are also used in agriculture for planting tasks as well as for keeping the application of pesticides within minimal amounts to mitigate environmental pollution. In this paper we present the development of a system to control the navigation of an autonomous mobile robot through tracks in plantations. Track images are used to control robot direction by pre-processing them to extract image features. Such features are then submitted to a support vector machine and an artificial neural network in order to find out the most appropriate route. A comparison of the two approaches was performed to ascertain the one presenting the best outcome. The overall goal of the project to which this work is connected is to develop a real time robot control system to be embedded into a hardware platform. In this paper we report the software implementation of a support vector machine and of an artificial neural network, which so far presented respectively around 93% and 90% accuracy in predicting the appropriate route. (C) 2013 The Authors. Published by Elsevier B.V. Selection and peer review under responsibility of the organizers of the 2013 International Conference on Computational Science

Distributes award protocol: A cooperation and communication method for robots.

In this Master’s Thesis a new Distributed Award Protocol (DAP) for robot communication and cooperation is presented. Task assignment (contract awarding) is done dynamically with contracts assigned to robots based upon the best bid received. Instead of having a manager and a contractor it is proposed a fully distributed bidding/awarding mechanism without a distinguished master. The best bidding robots are awarded with contract for execution. The contractors make decisions locally. This brings the following benefits: no communication bottleneck, low computational power requirement, increased robustness. DAP can handle multitasking. Tasks can be injected into system during the execution of already allocated tasks. As tasks have priorities, in the next cycle after taking into account actual bid parameters of all robots, tasks can be re-allocated. The aim is to minimize a global cost function which is a compromise between cost of task execution and cost of resources usage. Information about tasks and bid values is spread among robots with the use of a Round Robin Route, which is a novel solution proposed in this work. This method allows also identifying failed robots. Such failed robot is eliminated from the list of awarded robots and its replacement is found so the task is still executed by a team. If the failure of a robot was temporary (e.g. communication noise) and the robot can recover, it can again participate in the next bidding/awarding process. Using a bidding/awarding mechanism allows robots to dynamically relocate among tasks. This is also contributes to system robustness. DAP was evaluated through multiple experiments done in the multi-robot simulation system. Various scenarios were tested to check the idea of the main algorithm. Different failures of robots (communication failures, partial hardware malfunctions) were simulated and observations were made regarding how DAP recovers from them. Also the DAP flexibility to environment changes was watched. The experiments in the simulated environment confirmed the above features of DAP.

3DCOMET: 3D compression methods test dataset

The use of 3D data in mobile robotics applications provides valuable information about the robot’s environment. However usually the huge amount of 3D information is difficult to manage due to the fact that the robot storage system and computing capabilities are insufficient. Therefore, a data compression method is necessary to store and process this information while preserving as much information as possible. A few methods have been proposed to compress 3D information. Nevertheless, there does not exist a consistent public benchmark for comparing the results (compression level, distance reconstructed error, etc.) obtained with different methods. In this paper, we propose a dataset composed of a set of 3D point clouds with different structure and texture variability to evaluate the results obtained from 3D data compression methods. We also provide useful tools for comparing compression methods, using as a baseline the results obtained by existing relevant compression methods.

The application of robotics to the turbine blade encapsulation process

A survey of the existing state-of-the-art of turbine blade manufacture highlights two operations that have not been automated namely that of loading of a turbine blade into an encapsulation die, and that of removing a machined blade from the encapsulation block. The automation of blade decapsulation has not been pursued. In order to develop a system to automate the loading of an encapsulation die a prototype mechanical handling robot has been designed together with a computer controlled encapsulation die. The robot has been designed as a mechanical handling robot of cylindrical geometry, suitable for use in a circular work cell. It is the prototype for a production model to be called `The Cybermate'. The prototype robot is mechanically complete but due to unforeseen circumstances the robot control system is not available (the development of the control system did not form a part of this project), hence it has not been possible to fully test and assess the robot mechanical design. Robot loading of the encapsulation die has thus been simulated. The research work with regard to the encapsulation die has focused on the development of computer controlled, hydraulically actuated, location pins. Such pins compensate for the inherent positional inaccuracy of the loading robot and reproduce the dexterity of the human operator. Each pin comprises a miniature hydraulic cylinder, controlled by a standard bidirectional flow control valve. The precision positional control is obtained through pulsing of the valves under software control, with positional feedback from an 8-bit transducer. A test-rig comprising one hydraulic location pin together with an opposing spring loaded pin has demonstrated that such a pin arrangement can be controlled with a repeatability of +/-.00045'. In addition this test-rig has demonstrated that such a pin arrangement can be used to gauge and compensate for the dimensional error of the component held between the pins, by offsetting the pin datum positions to allow for the component error. A gauging repeatability of +/- 0.00015' was demonstrated. This work has led to the design and manufacture of an encapsulation die comprising ten such pins and the associated computer software. All aspects of the control software except blade gauging and positional data storage have been demonstrated. Work is now required to achieve the accuracy of control demonstrated by the single pin test-rig, with each of the ten pins in the encapsulation die. This would allow trials of the complete loading cycle to take place.

Role of reactive oxygen species in protein degradation in murine myotubes induced by proteolysis-inducing factor and angiotensin II

The antioxidants butylated hydroxytoluene (BHT, 1 mM) and d-α-tocopherol (10 μM) completely attenuated protein degradation in murine myotubes in response to both proteolysis-inducing factor (PIF) and angiotensin II (Ang II), suggesting that the formation of reactive oxygen species (ROS) plays an important role in this process. Both PIF and Ang II induced a rapid and transient increase in ROS formation in myotubes, which followed a parabolic dose-response curve, similar to that for total protein degradation. Antioxidant treatment attenuated the increase in expression and activity of the ubiquitin-proteasome proteolytic pathway by PIF and Ang II, by preventing the activation of the transcription factor nuclear factor-κB (NF-κB), through inhibition of phosphorylation of the NF-κB inhibitor protein (I-κB) and its subsequent degradation. ROS formation by both PIF and Ang II was attenuated by diphenyleneiodonium (10 μM), suggesting that it was mediated through the NADPH oxidase system. ROS formation was also attenuated by trifluoroacetyl arachidonic acid (10 μM), a specific inhibitor of cytosolic phospholipase A2, U-73122 (5 μM) and D609 (200 μM), inhibitors of phospholipase C and calphostin C (300 nM), a highly specific inhibitor of protein kinase C (PKC), all known activators of NADPH oxidase. Myotubes containing a dominant-negative mutant of PKC did not show an increase in ROS formation in response to either PIF or Ang II. The two Rac1 inhibitors W56 (200 μM) and NSC23766 (10 μM) also attenuated both ROS formation and protein degradation induced by both PIF and Ang II. Rac1 is known to mediate signalling between the phosphatidylinositol-3 kinase (PI-3K) product and NADPH oxidase, and treatment with LY24002 (10 μM), a highly selective inhibitor of PI-3K, completely attenuated ROS production in response to both PIF and Ang II, and inhibited total protein degradation, while the inactive analogue LY303511 (100 μM) had no effect. ROS formation appears to be important in muscle atrophy in cancer cachexia, since treatment of weight losing mice bearing the MAC16 tumour with d-α-tocopherol (1 mg kg- 1) attenuated protein degradation and increased protein synthesis in skeletal muscle. © 2007 Elsevier Inc. All rights reserved.

An optical fiber Bragg grating tactile sensor

Tactile sensors are needed for many emerging robotic and telepresence applications such as keyhole surgery and robot operation in unstructured environments. We have proposed and demonstrated a tactile sensor consisting of a fibre Bragg grating embedded in a polymer "finger". When the sensor is placed in contact with a surface and translated tangentially across it measurements on the changes in the reflectivity spectrum of the grating provide a measurement of the spatial distribution of forces perpendicular to the surface and thus, through the elasticity of the polymer material, to the surface roughness. Using a sensor fabricated from a Poly Siloxane polymer (Methyl Vinyl Silicone rubber) spherical cap 50 mm in diameter, 6 mm deep with an embedded 10 mm long Bragg grating we have characterised the first and second moment of the grating spectral response when scanned across triangular and semicircular periodic structures both with a modulation depth of 1 mm and a period of 2 mm. The results clearly distinguish the periodicity of the surface structure and the differences between the two different surface profiles. For the triangular structure a central wavelength modulation of 4 pm is observed and includes a fourth harmonic component, the spectral width is modulated by 25 pm. Although crude in comparison to human senses these results clearly shown the potential of such a sensor for tactile imaging and we expect that with further development in optimising both the grating and polymer "finger" properties a much increased sensitivity and spatial resolution is achievable.