Estratégias baseadas em aprendizado para coordenação de uma frota de robôs em tarefas cooperativas

In multi-robot systems, both control architecture and work strategy represent a challenge for researchers. It is important to have a robust architecture that can be easily adapted to requirement changes. It is also important that work strategy allows robots to complete tasks efficiently, considering that robots interact directly in environments with humans. In this context, this work explores two approaches for robot soccer team coordination for cooperative tasks development. Both approaches are based on a combination of imitation learning and reinforcement learning. Thus, in the first approach was developed a control architecture, a fuzzy inference engine for recognizing situations in robot soccer games, a software for narration of robot soccer games based on the inference engine and the implementation of learning by imitation from observation and analysis of others robotic teams. Moreover, state abstraction was efficiently implemented in reinforcement learning applied to the robot soccer standard problem. Finally, reinforcement learning was implemented in a form where actions are explored only in some states (for example, states where an specialist robot system used them) differently to the traditional form, where actions have to be tested in all states. In the second approach reinforcement learning was implemented with function approximation, for which an algorithm called RBF-Sarsa($lambda$) was created. In both approaches batch reinforcement learning algorithms were implemented and imitation learning was used as a seed for reinforcement learning. Moreover, learning from robotic teams controlled by humans was explored. The proposal in this work had revealed efficient in the robot soccer standard problem and, when implemented in other robotics systems, they will allow that these robotics systems can efficiently and effectively develop assigned tasks. These approaches will give high adaptation capabilities to requirements and environment changes.

Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

Viabilidade da Utilização de Máquinas de Indução Convencionais como Motores sem Mancais Mecânicos

Electrical Motors transform electrical energy into mechanic energy in a relatively easy way. In some specific applications, there is a need for electrical motors to function with noncontaminated fluids, in high speed systems, under inhospitable conditions, or yet, in local of difficult access and considerable depth. In these cases, the motors with mechanical bearings are not adequate as their wear give rise to maintenance. A possible solution for these problems stems from two different alternatives: motors with magnetic bearings, that increase the length of the machine (not convenient), and the bearingless motors that aggregate compactness. Induction motors have been used more and more in research, as they confer more robustness to bearingless motors compared to other types of machines building with others motors. The research that has already been carried out with bearingless induction motors utilized prototypes that had their structures of stator/rotor modified, that differ most of the times from the conventional induction motors. The goal of this work is to study the viability of the use of conventional induction Motors for the beringless motors applications, pointing out the types of Motors of this category that can be more useful. The study uses the Finite Elements Method (FEM). As a means of validation, a conventional induction motor with squirrel-cage rotor was successfully used for the beringless motor application of the divided winding type, confirming the proposed thesis. The controlling system was implemented in a Digital Signal Processor (DSP)

Algoritmos genéticos aplicados a um comitê de LS-SVM em problemas de classificação

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

Controle vetorial para velocidade de um motor de indução trifásico utilizando estimador filtro de Kalman

This paper describes the study, computer simulation and feasibility of implementation of vector control speed of an induction motor using for this purpose the Extended Kalman Filter as an estimator of rotor flux. The motivation for such work is the use of a control system that requires no sensors on the machine shaft, thus providing a considerable cost reduction of drives and their maintenance, increased reliability, robustness and noise immunity as compared to control systems with conventional sensors

Ambiente de desenvolvimentoweb multiplataforma configurável para robótica educacional

We propose an online configurable multiplatform development environment specifically developed for educational robotics applications. The environment, which appears as an extension of RoboEduc software, allows the programming of several programmable robots to be performed using the R-Educ language. We make it possible for the user to program in the language R-Educ and then translate the code to a language previously registered, compiled and then sent or executed by the robot. To develop this work, we conducted a bibliographic research about the main programming languages used in robotics, as well as their definitions and paradigms, from which it was possible to define a set of patterns considered important for the creation of this environment. Then, in the software development phase, we implemented the development environment, bearing in mind the requirements and functionality defined in the design phase. Finally, to validate the platform, we conducted some trials of programming languages and verificate if the complete cycle was satisfied - registration of language, programming in R-Educ, compilation for the registered language, compilation to the machine code and send the code for the robot

Implementação paralela escalável e eficiente do algoritmo simplex padrão em arquitetura multicore

This work presents a scalable and efficient parallel implementation of the Standard Simplex algorithm in the multicore architecture to solve large scale linear programming problems. We present a general scheme explaining how each step of the standard Simplex algorithm was parallelized, indicating some important points of the parallel implementation. Performance analysis were conducted by comparing the sequential time using the Simplex tableau and the Simplex of the CPLEXR IBM. The experiments were executed on a shared memory machine with 24 cores. The scalability analysis was performed with problems of different dimensions, finding evidence that our parallel standard Simplex algorithm has a better parallel efficiency for problems with more variables than constraints. In comparison with CPLEXR , the proposed parallel algorithm achieved a efficiency of up to 16 times better

Sistema inteligente para detecção de manchas de óleo na superfície marinha através de imagens de SAR

Oil spill on the sea, accidental or not, generates enormous negative consequences for the affected area. The damages are ambient and economic, mainly with the proximity of these spots of preservation areas and/or coastal zones. The development of automatic techniques for identification of oil spots on the sea surface, captured through Radar images, assist in a complete monitoring of the oceans and seas. However spots of different origins can be visualized in this type of imaging, which is a very difficult task. The system proposed in this work, based on techniques of digital image processing and artificial neural network, has the objective to identify the analyzed spot and to discern between oil and other generating phenomena of spot. Tests in functional blocks that compose the proposed system allow the implementation of different algorithms, as well as its detailed and prompt analysis. The algorithms of digital image processing (speckle filtering and gradient), as well as classifier algorithms (Multilayer Perceptron, Radial Basis Function, Support Vector Machine and Committe Machine) are presented and commented.The final performance of the system, with different kind of classifiers, is presented by ROC curve. The true positive rates are considered agreed with the literature about oil slick detection through SAR images presents

Desgaste de polímeros estruturais de engenharia em contato de deslizamento com cilindro metálico

Present work proposed to map and features the wear mechanisms of structural polymers of engineering derived of the sliding contact with a metallic cylindrical spindle submitted to eccentricity due to fluctuations in it is mass and geometric centers. For this it was projected and makes an experimental apparatus from balancing machine where the cylindrical counterbody was supported in two bearings and the polymeric coupon was situated in a holder with freedom of displacement along counterbody. Thus, the experimental tests were standardized using two position of the two bearings (Fixed or Free) and seven different positions along the counterbody, that permit print different conditions to the stiffness from system. Others parameters as applied normal load, sliding velocity and distance were fixed. In this investigation it was used as coupon two structural polymers of wide quotidian use, PTFE (polytetrafluroethylene) and PEEK (poly-ether-ether-ketone) and the AISI 4140 alloy steel as counterbody. Polymeric materials were characterized by thermal analysis (thermogravimetric, differential scanning calorimetry and dynamic-mechanical), hardness and rays-X diffractometry. While the metallic material was submitted at hardness, mechanical resistance tests and metallographic analysis. During the tribological tests were recorded the heating response with thermometers, yonder overall velocity vibration (VGV) and the acceleration using accelerometers. After tests the wear surface of the coupons were analyzed using a Scanning Electronic Microscopy (SEM) to morphological analysis and spectroscopy EDS to microanalysis. Moreover the roughness of the counterbody was characterized before and after the tribological tests. It was observed that the tribological response of the polymers were different in function of their distinct molecular structure. It were identified the predominant wear mechanisms in each polymer. The VGV of the PTFE was smaller than PEEK, in the condition of minimum stiffness, in function of the higher loss coefficient of that polymer. Wear rate of the PTFE was more of a magnitude order higher than PEEK. With the results was possible developed a correlation between the wear rate and parameter (E/ρ)1/2 (Young modulus, E, density, ρ), proportional at longitudinal elastic wave velocity in the material.

Estudo da flamabilidade e resistência à chama de compósito de poliéster insaturado e fibra da folha do abacaxizeiro (PALF)

In the present research work, composites were prepared using pine apple leaf fibres (PALF) as reinforcement with unsaturated polyester resin as matrix, incorporating with fire retardant at different compositions. The PALF was obtained from the decortication of pine apple leaves obtained from Ramada 4 from Ielmo Marinho in the State of Rio Grande do Norte. The unsaturated polyester resin and the catalyzer were bought from the local establishment. The fire retardant, aluminium tri-hydroxide - Al(OH)3 was donated by Alcoa Alumínio S.A and was used in the proportions of 20%, 40% and 60% w/w. Initially the fibres were treated with 2% NaOH for 1 hour, to remove any impurities present on the fibre surface, such as wax, fat, pectin and pectate, in order to have a better adsorption of the fibres with the matrix as well as the flame retardant. The fibre mat was prepared in a mat preparator by immersion, developed in the Textile Engineering Laboratory, at the UFRN. The composites (300x300x3 mm) were prepared by compression molding and the samples (150x25x3 mm) for analysis of the properties were cut randomly using a laser cutter. Some of the cut samples were used to measure the smoke emission and fire resistance using UL94 standard. Mechanical tension-extension and flexural properties were carried in CTGás RN and the Laboratório de Metais e Ensaios Mecânicos Engenharia de Materiais UFRN , as well as SEM studies were carried out at Núcleo de Estudos em Petróleo e Gás Natural - UFRN . From the observed results, it was noted that, there was no marked influence of the fire retardant on the mechanical properties. Also in the water absorption test, the quantity of water absorbed was less in the sample with higher concentration of fire retardant. It was also observed that the increase in the proportion of the fire retardant increased the time of burning, may be due to the compactness of the composite due to the presence of fire retardant as a filling material even though it was meant to reduce the rate of inflammability of the composite

Desenvolvimento de uma desfribadeira para obtenção da fibra da folha do abacaxi

In 1998 the first decorticator was developed in the Textile Engineering Laboratory and patented for the purpose of extracting fibres from pineapple leaves, with the financial help from CNPq and BNB. The objective of the present work was to develop an automatic decorticator different from the first one with a semiautomatic system of decortication with automatic feeding of the leaves and collection of the extracted fibres. The system is started through a command system that passes information to two engines, one for starting the beater cylinder and the other for the feeding of the leaves as well as the extraction of the decorticated fibres automatically. This in turn introduces the leaves between a knife and a beater cylinder with twenty blades (the previous one had only 8 blades). These blades are supported by equidistant flanges with a central transmission axis that would help in increasing the number of beatings of the leaves. In the present system the operator has to place the leaves on the rotating endless feeding belt and collect the extracted leaves that are being carried out through another endless belt. The pulp resulted form the extraction is collected in a tray through a collector. The feeding of the leaves as well as the extraction of the fibres is controlled automatically by varying the velocity of the cylinders. The semi-automatic decorticator basically composed of a chassis made out of iron bars (profile L) with 200cm length, 91 cm of height 68 cm of width. The decorticator weighs around 300Kg. It was observed that the increase in the number of blades from 8 to twenty in the beater cylinder reduced the turbulence inside the decorticator, which helped to improve the removal of the fibres without any problems as well as the quality of the fibres. From the studies carried out, from each leaf 2,8 to 4,5% of fibres can be extracted. This gives around 4 to 5 tons of fibres per hectare, which is more than that of cotton production per hectare. This quantity with no doubt could generate jobs to the people not only on the production of the fibres but also on their application in different areas

Compósito reforçado com sisal para aplicação na moda

The objective of this research is the fabrication of a composite reinforced with dyed sisal fiber and polyester matrix for application in the fields such as, fashion, clothing, interior textiles; fashion accessories are some of the examples. For the fabrication of the composite, the sisal fibers were subjected to processes such as: chemical treatment with sodium hydroxide (NaOH) in the removal of impurities; bleaching for removing the yellowish color of the natural fiber and dyeing with direct dyes to confer the colors blue, green and orange. The search for new technologies ecologically correct has become a major concern in recent decades. Studies show that composite polymer reinforced by natural fibers is suitable for a large number of applications, and its use is advantageous in terms of economic and ecological. The dyed fibers were cut to a length of 30 mm, is used in the confection of webs. For this purpose, a web preparer by immersion, developed in the Laboratory of Chemical Textile of UFRN. The composite sheets measuring 300 x 300 x3 mm were molded by compression, with unsaturated orthophthalic polyester as matrix, and the samples in sizes 150 x 25 x 3 mm were cut with the aid of a laser machine, to be subjected to traction and flexion. The mechanical properties of traction and flexion in three points were performed in the Laboratory of metal and mechanical tests of Materials Engineering of UFRN. The resulting samples from the tests were evaluated in scanning electron microscope (SEM) at CTGas RN. On the basis of the analysis of the results from the mechanical tests, it was observed that the composite had good mechanical behavior, both in traction as in flexion. Furthermore, it was observed that in the water absorption test, the samples had a different percentage among themselves, this occurred due to the variation of density found in the fibre webs. The images of the SEM showed the failures from the manufacturing process and the adhesion of fibre/matrix. When the samples were prepared with the dyed fibers to be applied in fashion, the results were positive, and it can be concluded that the main objective of this work was achieved

Beneficiamento de sementes de café e efeitos na qualidade física

A pesquisa foi realizada com o objetivo de verificar a possibilidade de uso da máquina de pré-limpeza, da máquina de limpeza e classificação (ventiladores e peneiras) e da mesa densimétrica para o beneficiamento de sementes de café e, paralelamente, avaliar seus efeitos na qualidade física do lote. Para essa finalidade, um lote de sementes de café arábica 'Catuaí Amarelo IAC 62' foi submetido ao beneficiamento nesses três equipamentos, em diversas combinações. As frações retidas pelas peneiras 22/64 e 20/64 avos de polegada, na máquina de ventiladores e peneiras, constituíram a maior parte do lote e foram processadas, separadamente, em mesa densimétrica. Foram realizadas avaliações de pureza física, massa de mil sementes, peso hectolítrico, teor de água, tipos de sementes predominantes e integridade do endocarpo. Observou-se que: a) A máquina de pré-limpeza, a máquina de ventiladores e peneiras e a mesa densimétrica foram eficientes para a homogeneização do lote de sementes de café, proporcionando obtenção de melhor qualidade física na seqüência operacional em que foram utilizadas de forma conjugada; b) O uso de peneiras com crivos oblongos e crivos circulares, intercaladas na máquina de ventiladores e peneiras, possibilitou a separação de sementes mocas, de sementes com resíduos de exocarpo aderidos ao endocarpo e de sementes desprovidas de endocarpo, melhorando a aparência do lote; c) A mesa densimétrica foi o equipamento que proporcionou melhor separação de sementes com endocarpo danificado, bem como alterações favoráveis à qualidade física do lote; d) As sementes pequenas, retidas nas menores peneiras da máquina de ventiladores e peneiras, e as sementes de menor massa específica, classificadas ou não por tamanho e coletadas na descarga inferior da mesa densimétrica, foram as de baixa qualidade física.

Beneficiamento de sementes de café e efeitos na qualidade fisiológica

Considerando-se a escassez de informações sobre o beneficiamento mecanizado de sementes de café, realizou-se este trabalho para estudar seus efeitos na qualidade fisiológica das sementes de café arábica. Um lote de sementes de 'Catuaí Amarelo IAC 62' foi submetido à ação de máquina de pré-limpeza, máquina de ventiladores e peneiras e mesa densimétrica, em diversas combinações, e as sementes foram avaliadas quanto à germinação e ao vigor (envelhecimento acelerado, emergência de plântulas, altura e massa de matéria seca de plântulas e mudas). O teor de água das sementes também foi determinado. Utilizou-se o delineamento estatístico de blocos ao acaso, com quatro repetições, e o teste de Tukey a 5% para comparação de médias. Concluiu-se que: a) a qualidade fisiológica das sementes foi aprimorada à medida que o lote passou por equipamentos específicos no beneficiamento, culminando com a melhor qualidade na seqüência operacional em que o lote foi submetido à ação da máquina de ventiladores e peneiras e mesa densimétrica; b) a mesa densimétrica, associada à máquina de ventiladores e peneiras, foi o equipamento que proporcionou a melhor separação de materiais que interferem negativamente na qualidade fisiológica das sementes; c) as sementes maiores e/ou mais pesadas tiveram qualidade fisiológica superior, e as sementes mocas graúdas, qualidade fisiológica similar à das sementes chatas graúdas; d) o descarte das frações do lote constituídas por sementes pequenas ou de menor densidade possibilitou a obtenção de lotes homogêneos e com alta qualidade fisiológica, favorecendo a obtenção de mudas mais vigorosas.

Distribuição longitudinal e germinação de sementes de milho com emprego de tratamento fitossanitário e grafite

Teve-se o objetivo de avaliar as formas de se misturar o lubrificante sólido grafite nas sementes de milho, as dosagens e a influência desse na distribuição longitudinal e na germinação, em sementes sem e com tratamento fitossanitário, em mecanismo dosador com disco perfurado horizontal. Os fatores analisados foram: distribuição longitudinal das sementes (espaçamentos falhos, múltiplos e aceitáveis) e a porcentagem de germinação aos 7 e aos 14 dias após a instalação no germinador. Concluiu-se que o tratamento fitossanitário aumentou os espaçamentos falhos e múltiplos e reduziu os espaçamentos aceitáveis. O emprego de grafite reduziu os espaçamentos falhos e múltiplos e elevou os espaçamentos aceitáveis. A melhor forma de mistura do grafite ocorreu na máquina de tratamento fitossanitário de sementes. A distribuição longitudinal alcançou seu ponto ótimo com a dosagem de grafite de 3,37 g kg-1 de semente. As variáveis tratamento fitossanitário, mecanismo dosador e grafite não influíram na germinação das sementes de milho.