Submitted by Setor de Tratamento da Informa??o - BC/PUCRS (tede2@pucrs.br) on 2016-09-27T12:01:23Z No. of bitstreams: 1 DIS_MARCELO_VASCONCELLOS_GOMES_COMPLETO.pdf: 2508819 bytes, checksum: 8ffeedd7b3c2ed3f9322ffff87185a34 (MD5) / Made available in DSpace on 2016-09-27T12:01:23Z (GMT). No. of bitstreams: 1 DIS_MARCELO_VASCONCELLOS_GOMES_COMPLETO.pdf: 2508819 bytes, checksum: 8ffeedd7b3c2ed3f9322ffff87185a34 (MD5) Previous issue date: 2016-06-30 / This study presents a method to evaluate the performance of teams of software development projects using Stochastic Automata Networks. For the application of this method, a ?Performance Evaluation? tool was created. This tool can be used by Project Managers and Metrics Analysts to simulate scenarios of execution in the projects. According to the present performance of the project team, the Project Manager can foresee the necessary effort to accomplish the project, probable date for its conclusion and its total cost.Through the result of this simulation, the Project Manager will be able to take the necessary actions to mitigate the impact in the project deadlines and cost. Furthermore, the Metrics Analyst can validate the best productivity to be used in the project. The tool has a great potential to be used together with good project management practices described in the PMBoK Guide. The study describes all the processes of the PMBoK with focus in the group of monitoring and control processes and how the Performance Evaluation tool can contribute to the project management. The study also presents related works where Stochastic Automata Networks can contribute significantly to Software Engineering area. The study also presents the evaluation of two scenarios that were created in the tool using real data of two projects and in the end a comparison was made between the real data and the simulation results. Finally, improvements and suggestions are presented for future implementations in the ?Performance Evaluation? tool. / Este estudo apresenta um m?todo para avaliar o desempenho de equipes de projetos de desenvolvimento de software usando Redes de Aut?matos Estoc?sticos. Para a aplica??o deste m?todo foi criada a ferramenta ?Avalia??o de Desempenho?. Esta ferramenta pode ser usada por Gerentes de Projetos e Analistas de M?tricas para simular cen?rios de execu??o dos projetos. De acordo com o desempenho atual da equipe do projeto, o Gerente de Projetos poder? prever o esfor?o necess?rio para realizar o projeto, a data prov?vel para a sua conclus?o e o seu custo total. Atrav?s do resultado desta simula??o, o Gerente de Projetos ser? capaz de tomar as a??es necess?rias para mitigar o impacto no prazo e custo do projeto. Al?m disso, o Analista de M?tricas poder? validar a melhor produtividade a ser usada no projeto. A ferramenta tem um grande potencial para ser usada em conjunto com as boas pr?ticas de gest?o de projetos descritas no Guia do PMBoK. O estudo descreve todos os processos do PMBoK com foco no grupo de processos de monitoramento e controle e como a ferramenta de avalia??o de desempenho pode contribuir na gest?o do projeto. O estudo tamb?m apresenta trabalhos relacionados onde Redes de Aut?matos Estoc?sticos podem contribuir significativamente para a ?rea da Engenharia de Software. O estudo tamb?m apresenta a avalia??o de dois cen?rios que foram criados na ferramenta com dados reais de dois projetos e no final foi realizada uma compara??o entre os dados reais e os resultados da simula??o. Por fim, s?o apresentadas melhorias e sugest?es para futuras implementa??es na ferramenta ?Avalia??o de Desempenho?.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-04-07T20:43:48Z No. of bitstreams: 1 FabioBarretoRodriguesDissertacao2015.pdf: 3967415 bytes, checksum: f7cf99ba46734ee5644ff6b6f3b44749 (MD5) / Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-04-07T20:44:08Z (GMT) No. of bitstreams: 1 FabioBarretoRodriguesDissertacao2015.pdf: 3967415 bytes, checksum: f7cf99ba46734ee5644ff6b6f3b44749 (MD5) / Made available in DSpace on 2017-04-07T20:44:08Z (GMT). No. of bitstreams: 1 FabioBarretoRodriguesDissertacao2015.pdf: 3967415 bytes, checksum: f7cf99ba46734ee5644ff6b6f3b44749 (MD5) Previous issue date: 2015-07-16 / The reduced dopamine production in Parkinson's disease (PD) substantially affects the motor skills and other primary bodily functions, including mood, cognition and functional performance. At the same time, losses occur in the autonomic nervous system. It is unknown how the autonomic function is related to these losses and whether a walking program can modify the heart rate variability (HRV) in this population. The aim of this study was to assess HRV in patients with PD and its relationship with anxiety and depression symptons, functional performance and severity of disease, and also the effect of a walking program in autonomic modulation and functional capacity. In cross- sectional phase were evaluated 26 patients (16 men / 10 women, 57.4 ?? 9.2 years) using the scales of anxiety, depression (HADS-A and HADS-D) and disease severity (UPDRS motorexamination and Hoehn and Yahrd scales). Functional performance was evaluated by the Timed Up and Go test (TUG) and three tests from the Senior Fitness Tests. These results were correlated with HRV indexes of time (RMSSD, SDNN) and frequency domain (HF, LF, LF / HF, TP, HF / TP, LF / TP) and nonlinear analysis (SampEn and DFA1) measured in sitting and standing position. The HRV variation from sitting to standing position (??? analysis) was also considered. In longitudinal phase a walking group (WG, n = 13) and a control group (CG n = 6) were formed through a randomized controlled trial. The home-based walking program had a duration of six weeks. The patients in the WG group were reassessed after 12 weeks of training. Were observed high levels of anxiety (30.7%) and depression (34.6%) in the patients evaluated. Depressive symptoms were associated to the parasympathetic activity: RMSSD (r = - 0.382; p ??? 0.05), as the activity in the both sympathetic and parasympathetic branches: LFseated (r = -0383; p ??? 0.05), LFstanding (r = -0.411; p ??? 0.05), TPstanding (r = -0, 504, p ??? 0.05) and ??? SampEn (r = -0384, p ??? 0.05). The severity and duration of illness also were negatively associated with autonomic function. RMSSDseated (r = -0.438; p ??? 0.05), SDNNseated (r = -0.452; p ??? 0.05) SDNNstanding (r = -0.602, p ??? 0.01); LFstanding (r = -0.603; p ??? 0.01); TPstanding (r = -0.504; p ??? 0.05) and LF/ TPstanding (r = -0.504; p ??? 0.05) were negatively associated with time to perform the TUG test. Other significant correlations (p ??? 0.05) of physical and functional tests and age HRV were observed . The six-weeks walking program have promoted significant changes in SampEn index (which measures the irregularity of HR data) in WG group compared to the CG group (20.47 vs -22.81%; CA and CO, respectively; p = 0.033) as well as ???SampEn measure (-22.83% vs. 18.46; WG and CG, respectively; p = 0.022), suggesting positive changes in autonomic function. At week 12, the WG group patients exhibited increased ratios of HF/TPstanding (7.0%; p = 0.017) and SampEn-standing (20.5%; p = 0.008; respectively), and reduction of the LF/HFstanding index (-10.65%; p = 0.012 ) compared to the first week, suggesting improvements in autonomic balance. There was an increase in sit-to-stand repetitions (10.50 ?? 1.93 to 13.25 ?? 1.58 repetitions; p = 0.021) and reduced time to perform the TUG test (7.34"?? 0.67 to 6.91" ?? 0.83, p = 0.002). Several HRV indices measured before walking program (SDNN, RMSSD, LF, HF, LF / TP) have shown correlations with the functional gain (r2 ranging from 0.62 to 0.81), suggesting a role of autonomic function in neuromuscular adaptation to training. It follows that HRV may work as an integrative measure and also as a peripheral marker of body's adaptability and its self-regulatory capacity, besides a practical and sensitive tool e to the disease progression. The home-based walking program have improved autonomic function of participants, often hampered by the disease. This is particularly interesting in that many bodily functions impaired in PD are mediated by autonomic function. / A redu????o da produ????o de dopamina na Doen??a de Parkinson (DP) afeta substancialmente a motricidade e outras fun????es org??nicas primordiais, como humor e a cogni????o, bem como o desempenho funcional. Paralelamente, ocorrem preju??zos no sistema nervoso aut??nomo. N??o se sabe como a fun????o auton??mica se relaciona com estes preju??zos e se um programa de caminhada pode modificar a variabilidade da frequ??ncia card??aca (VFC) desta popula????o. O objetivo deste estudo foi avaliar a VFC de pacientes com DP e a sua rela????o com os sintomas de ansiedade, depress??o, desempenho funcional e gravidade da doen??a, e tamb??m o efeito de um programa de caminhada na modula????o auton??mica e na capacidade funcional. Na etapa transversal foram avaliados 26 pacientes (16 homens/ 10 mulheres; 57,4 ?? 9,2 anos) pelas escalas de ansiedade, depress??o (HADS-A e HADS-D) e gravidade da doen??a (UPDRSexame motor e Hoehn e Yahrd). O desempenho funcional foi avaliado pelos testes Timed up and Go (TUG) e tr??s testes da bateria Senior Fitness Test. Estes resultados foram correlacionados com ??ndices da VFC no dom??nio do tempo (RMSSD, SDNN) e da frequ??ncia (HF, LF, LF/HF, TP, HF/TP, LF/TP), e an??lise n??o linear (SampEn e DFA1) aferidos na posi????o sentado e em p??. A varia????o da VFC da posi????o sentada para em p?? (valores em ??) tamb??m foi considerada. Na etapa longitudinal foram formados um grupo de caminhada (CA; n=13) e um grupo controle (CO; n=6), a partir de um ensaio cl??nico randomizado. O programa de caminhada home-based teve dura????o de seis semanas. Os pacientes do grupo CA foram reavaliados ap??s 12 semanas de treinamento. Foram observados elevados n??veis de ansiedade (30,7%) e depress??o (34,6%) nos pacientes avaliados. Os sintomas depressivos associaram-se tanto ?? atividade parassimp??tica: RMSSD (r= - 0,382; p ??? 0,05), quanto ?? atividade conjunta nos ramos simp??tico e parassimp??tico: LFsent (r= -0.383; p ??? 0,05), LFem p?? (r= -0,411; p ??? 0,05), TPemp?? (r= -0, 504; p ??? 0,05) e ?? SampEn (r= -0.384; p ??? 0,05). A gravidade e tempo da doen??a tamb??m se associaram negativamente ?? fun????o auton??mica. RMSSDsentado (r= -0,438; p ??? 0,05), SDNNsentado (r= -0,452; p ??? 0,05) e SDNNemp?? (r= -0,602; p ??? 0,01); LFemp?? (r= -0,603; p ??? 0,01); TPemp?? (r= -0,504; p ??? 0,05) e LF/TPemp?? (r= -0,504; p ??? 0,05) foram negativamente associados ao tempo no teste TUG. Outras correla????es significativas dos testes f??sico-funcionais e da idade com a VFC foram observadas. O programa de seis semanas de caminhada promoveu varia????es significativas do ??ndice SampEn (que mede a irregularidade dos dados de FC) no grupo CA em rela????o ao grupo CO (20,47 vs ???22,81%; CA e CO, respectivamente; p=0,033), assim como o ??ndice ??SampEn (-22,83 vs 18,46%; CA e CO, respectivamente; p=0,022), sugerindo mudan??as positivas na fun????o auton??mica. Na 12?? semana, os pacientes do grupo CA exibiram aumento dos ??ndices HF/TPemp?? (7,0%; p=0,017) e SampEn-em p?? (20,5%, p= 0,008), respectivamente) e redu????o do ??ndice LF/HFemp?? (-10,65 %; p=0,012) em rela????o ?? 1?? semana, sugerindo melhoras no equil??brio auton??mico. Houve aumento no n??mero de agachamentos (de 10,50 ?? 1,93 para 13,25 ?? 1,58 repeti????es; p= 0,021) e redu????o no tempo para realizar o TUG ( de 7,34??? ?? 0,67 para 6,91 ?? 0,83, p=0,002). V??rios ??ndices de VFC aferidos na tes do programa de caminhada (SDNN, RMSSD, LF, HF, LF/TP) mostraram correla????es com o ganho funcional (r2 variando de 0,62 a 0,81), sugerindo papel da fun????o auton??mica na adapta????o neuromuscular ao treinamento. Conclui-se que a VFC pode funcionar como medida integrativa e marcador perif??rico da adaptabilidade do organismo e da sua capacidade autoregulat??ria, al??m de recurso de avalia????o pr??tico e sens??vel ?? progress??o da doen??a. O programa de caminhada home-based melhorou a fun????o auton??mica dos participantes, frequentemente prejudicada pela doen??a. Isto ?? particularmente interessante na medida em que v??rias fun????es org??nicas debilitadas na DP s??o mediadas pela fun????o auton??mica.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-08-16T20:20:26Z No. of bitstreams: 1 AtilaVarelaFerreiraMedeirosDeOliveira_DISSERT.pdf: 1603393 bytes, checksum: dca520d8a49b825f11ed063f8807d7f1 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-08-18T00:33:44Z (GMT) No. of bitstreams: 1 AtilaVarelaFerreiraMedeirosDeOliveira_DISSERT.pdf: 1603393 bytes, checksum: dca520d8a49b825f11ed063f8807d7f1 (MD5) / Made available in DSpace on 2016-08-18T00:33:44Z (GMT). No. of bitstreams: 1 AtilaVarelaFerreiraMedeirosDeOliveira_DISSERT.pdf: 1603393 bytes, checksum: dca520d8a49b825f11ed063f8807d7f1 (MD5) Previous issue date: 2015-12-07 / Este trabalho prop?e uma nova estrat?gia de navega??o aut?noma assistida por algoritmo gen?tico com planejamento din?mico para rob?s m?veis terrestres, chamada DPNA-GA (Dynamic Planning Navigation Algorithm optimized with Genetic Algorithm). A estrat?gia foi aplicada em ambientes - tanto est?ticos, quanto din?micos - nos quais a localiza??o e o formato dos obst?culos n?o s?o previamente conhecidos. A cada evento de deslocamento, uma nova rota ? planejada atrav?s de um algoritmo que minimiza a dist?ncia entre o rob? e o objetivo e maximiza a dist?ncia em rela??o aos obst?culos. Utilizando um sensor de localiza??o espacial e um conjunto de sensores de dist?ncia, a estrat?gia de navega??o proposta foi capaz de planejar dinamicamente percursos ?timos livres de colis?o. Simula??es realizadas em diferentes ambientes demostraram que a t?cnica fornece um alto grau de flexibilidade e robustez. Para isso, foram aplicadas diversas varia??es de par?metros gen?ticos, tais como: taxa de cruzamento, tamanho da popula??o, dentre outros. Finalmente, os resultados das simula??es demonstram satisfatoriamente a efic?cia e robustez da t?cnica DPNA-GA, validando-a para aplica??es reais em rob?s m?veis terrestres. / This work proposes a new autonomous navigation strategy assisted by genetic algorithm with dynamic planning for terrestrial mobile robots, called DPNA-GA (Dynamic Planning Navigation Algorithm optimized with Genetic Algorithm). The strategy was applied in environments - both static and dynamic - in which the location and shape of the obstacles is not known in advance. In each shift event, a control algorithm minimizes the distance between the robot and the object and maximizes the distance from the obstacles, rescheduling the route. Using a spatial location sensor and a set of distance sensors, the proposed navigation strategy is able to dynamically plan optimal collision-free paths. Simulations performed in different environments demonstrated that the technique provides a high degree of flexibility and robustness. For this, there were applied several variations of genetic parameters such as: crossing rate, population size, among others. Finally, the simulation results successfully demonstrate the effectiveness and robustness of DPNA-GA technique, validating it for real applications in terrestrial mobile robots.

Submitted by Jos? Henrique Henrique (jose.neves@ufvjm.edu.br) on 2017-06-22T22:29:53Z No. of bitstreams: 2 everton_oliveira_paiva.pdf: 3688473 bytes, checksum: 00c67bcc4d4564b69bb64a0b596743fc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2017-06-23T13:21:09Z (GMT) No. of bitstreams: 2 everton_oliveira_paiva.pdf: 3688473 bytes, checksum: 00c67bcc4d4564b69bb64a0b596743fc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-06-23T13:21:09Z (GMT). No. of bitstreams: 2 everton_oliveira_paiva.pdf: 3688473 bytes, checksum: 00c67bcc4d4564b69bb64a0b596743fc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016 / O uso sistemas computacionais como complemento ou substitui??o da sala de aula ? cada vez mais comum na educa??o e os Sistemas Tutores Inteligentes (STIs) s?o uma dessas alternativas. Portanto ? fundamental desenvolver STIs capazes tanto de ensinar quanto aprender informa??es relevantes sobre o aluno atrav?s de t?cnicas de intelig?ncia artificial. Esse aprendizado acontece por meio da intera??o direta entre o STI e o aluno que ? geralmente demorada. Esta disserta??o apresenta a inser??o da metaheur?sticas Lista Tabu e GRASP com o objetivo de acelerar esse aprendizado. Para avaliar o desempenho dessa modifica??o, foi desenvolvido um simulador de STI. Nesse sistema, foram realizadas simula??es computacionais para comparar o desempenho da tradicional pol?tica de explora??o aleat?ria e as metaheur?sticas propostas Lista Tabu e GRASP. Os resultados obtidos atrav?s dessas simula??es e os testes estat?sticos aplicados indicam fortemente que a introdu??o de meta-heur?sticas adequadas melhoram o desempenho do algoritmo de aprendizado em STIs. / Disserta??o (Mestrado Profissional) ? Programa de P?s-Gradua??o em Educa??o, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2016. / Using computer systems as a complement or replacement for the classroom experience is an increasingly common practice in education and Intelligent Tutoring Systems (ITS) are one of these alternatives. Therefore, it is crucial to develop ITS that are capable of both teaching and learning relevant information about the student through artificial intelligence techniques. This learning process occurs by means of direct, and generally slow, interaction between the ITS and the student. This dissertation presents the insertion of meta-heuristic Tabu search and GRASP with the purpose of accelera ting learning. An ITS simulator was developed to evaluate the performance of this change. Computer simulations were conducted in order to compare the performance of traditional randomized search methods with the meta-heuristic Tabu search. Results obtained from these simulations and statistical tests strongly indicate that the introduction of meta-heuristics in exploration policy improves the performance of the learning algorithm in ITS.

Made available in DSpace on 2014-12-17T15:03:33Z (GMT). No. of bitstreams: 1 SalatielRV_DISSERT.pdf: 3344605 bytes, checksum: 083e5071106853d45a5c3fde7712fcac (MD5) Previous issue date: 2014-03-10 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / The Pitimbu River Watershed (PRW), belonging to Potiguar capital metropolitan area, State of Rio Grande do Norte, contributes, among other purposes, to human using and animal watering. This watershed is extremely important because, besides filling up with freshwater approximately 30% of the south part of Natal (South, East and West Zones), contributes to the river shore ecosystem equilibrium. Face to the current conjuncture, this study aims to evaluate the urban development dynamics in the PRW, applying Cellular Automata as a modeling instrument, and to simulate future urban scenarios, between 2014 and 2033, using the simulation program SLEUTH. In the calibration phase, urban spots for 1 984, 1992, 2004 and 2013 years were used, with resolution from 100 meters. After the simulation, it was found a predominance of organic growth, expanding the BHRP from existing urban centers. The spontaneous growth occurred through the fullest extent of the watershed, however the probability of effective growth should not exceed 21%. It was observed that, there was a 68% increase for the period between 2014 and 2033, corresponding to an expansion area of 1,778 ha. For 2033, the source of Pitimbu River area and the Jiqui Lake surroundings will increase more than 78%. Finally, it was seen an exogenous urban growth tendency in the watershed (outside-in). As a result of this growth, hydraulics resources will become scarcer / A Bacia Hidrogr?fica do Rio Pitimbu (BHRP), pertencente ? regi?o metropolitana da capital Potiguar, Estado do Rio Grande do Norte (RN), contribui, entre outros fins, para o consumo humano e dessedenta??o animal. Essa bacia ? de suma import?ncia, pois al?m de abastecer com ?gua doce aproximadamente 30% da popula??o da parte sul de Natal (zonas sul, leste e oeste), contribui para o equil?brio do ecossistema ao longo do rio. Diante da conjuntura atual, os objetivos deste estudo foram avaliar a din?mica do desenvolvimento urbano na BHRP, aplicando Aut?matos Celulares como instrumento de modelagem, e simular cen?rios urbanos futuros, entre 2014 e 2033, empregando o programa de simula??o SLEUTH. Na fase de calibra??o, foram utilizadas as manchas urbanas para os anos de 1984, 1992, 2004 e 2013, com resolu??o 100 metros. Ap?s a simula??o, verificou-se que houve uma predomin?ncia do crescimento org?nico, expandindo-se na BHRP, a partir de centros urbanos existentes. O crescimento espont?neo ocorreu por toda extens?o da Bacia, por?m a probabilidade de crescimento efetivo n?o deve ultrapassar 21%. Verificou-se um crescimento de 68% para o per?odo entre 2014 e 2033, correspondendo a uma ?rea de expans?o de 1.778 ha. Para o ano de 2033, a ?rea da nascente do rio Pitimbu e proximidades da lagoa do Jiqui ter?o a possibilidade efetiva de crescimento acima de 78%. Por fim, observou -se uma tend?ncia de crescimento urbano ex?geno (de fora para dentro) na Bacia. Em consequ?ncia desse crescimento, os recursos h?dricos tornar-se-?o cada vez mais escassos

This diploma thesis describes the design of showroom and car service of Mazda brand in Trencin. It is a detached building on a flat terrain. The car showroom and car service have two above-ground floors and one underground floor. On the first underground floor there is an underground garage designed for new cars for sale. On the first floor there is a car showroom with a customer base and a car service with facilities for car service staff. On the second floor there are offices with facilities for staff. The building is designed from a reinforced concrete structure and porous concrete blocks. The ceilings are monolithic reinforced concrete and the roof is designed as flat with trusses and reinforced concrete slab. The work includes project documentation for the construction.

The purpose of this diploma thesis „Parking Planning and Calculation„ is car parks at selected buildings. The goal of this thesis is to optimize the selected of targeted units for calculate the number of parking spaces. First part of the work is focused on the measurement itself and data analysis. The next part of the work consists of comparison of calculated parking places according to standards with real number of parking places.

The third platform and the center of sustainable development in Lausanne Sustainable city is in my opinion city, that offers to it's inhabitants good-quality city spaces – attractive streets and squares - interesting for spend free time. Main reason of this I see in insufficient regulation of car traffic. Big amount of cars takes big part of cities and burden it by dust and noise. Also the scale of cities is adapted for cars. That is way I designed city quarter without possibility of private car access. It's alternative to common ways and under arrest to the center of sustainable development it demonstrates it's preferences and becomes alive exhibit. In hart of third platform is situated former cargo railway-station Gare de Sébeillon. It designed for reconversion to the center of sustainable development. Architectural proposition is answered by urban concept.

Made available in DSpace on 2015-03-03T15:08:47Z (GMT). No. of bitstreams: 1 BrunoSJC_TESE.pdf: 2605632 bytes, checksum: cc7fdbd9d8d7dfe3adac23f17fab1ae2 (MD5) Previous issue date: 2014-05-13 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / In this work, we propose a two-stage algorithm for real-time fault detection and identification of industrial plants. Our proposal is based on the analysis of selected features using recursive density estimation and a new evolving classifier algorithm. More specifically, the proposed approach for the detection stage is based on the concept of density in the data space, which is not the same as probability density function, but is a very useful measure for abnormality/outliers detection. This density can be expressed by a Cauchy function and can be calculated recursively, which makes it memory and computational power efficient and, therefore, suitable for on-line applications. The identification/diagnosis stage is based on a self-developing (evolving) fuzzy rule-based classifier system proposed in this work, called AutoClass. An important property of AutoClass is that it can start learning from scratch". Not only do the fuzzy rules not need to be prespecified, but neither do the number of classes for AutoClass (the number may grow, with new class labels being added by the on-line learning process), in a fully unsupervised manner. In the event that an initial rule base exists, AutoClass can evolve/develop it further based on the newly arrived faulty state data. In order to validate our proposal, we present experimental results from a level control didactic process, where control and error signals are used as features for the fault detection and identification systems, but the approach is generic and the number of features can be significant due to the computationally lean methodology, since covariance or more complex calculations, as well as storage of old data, are not required. The obtained results are significantly better than the traditional approaches used for comparison / Este trabalho prop?e um algoritmo de dois estagios para detec??o e identifica??o de falhas, em tempo real, em plantas industriais. A proposta baseia-se na analise de caracter?sticas selecionadas utilizando estimativa de densidade recursiva e um novo algoritmo evolutivo de classifica??o. Mais especificamente, a abordagem proposta para detec??o e baseada no conceito de densidade no espa?o de dados, o que difere da tradicional fun??o densidade de probabilidade, porem, sendo uma medida bastante util na detec??o de anormalidades/outliers. Tal densidade pode ser expressa por uma fun??o de Cauchy e calculada recursivamente, o que torna o algoritmo computacionalmente eficiente, em termos de processamento e memoria, e, dessa maneira, apropriado para aplica??es on-line. O estagio de identifica??o/diagnostico e realizado por um classificador baseado em regras fuzzy capaz de se auto-desenvolver (evolutivo), chamado de AutoClass, e introduzido neste trabalho. Uma propriedade importante do AutoClass e que ele e capaz de aprender a partir do zero". Tanto as regras fuzzy, quanto o numero de classes para o algoritmo n?o necessitam de pre-especifica??o (o numero de classes pode crescer, com os rotulos de classe sendo adicionados pelo processo de aprendizagem on-line), de maneira n~ao-supervisionada. Nos casos em que uma base de regras inicial existe, AutoClass pode evoluir/desenvolver-se a partir dela, baseado nos dados adquiridos posteriormente. De modo a validar a proposta, o trabalho apresenta resultados experimentais de simula??o e de aplica??es industriais reais, onde o sinal de controle e erro s?o utilizados como caracter?sticas para os estagios de detec??o e identifica??o, porem a abordagem e generica, e o numero de caracter?sticas selecionadas pode ser significativamente maior, devido ? metodologia computacionalmente eficiente adotada, uma vez que calculos mais complexos e armazenamento de dados antigos n?o s?o necess?rios. Os resultados obtidos s?o signifificativamente melhores que os gerados pelas abordagens tradicionais utilizadas para compara??o

Made available in DSpace on 2014-12-17T14:56:04Z (GMT). No. of bitstreams: 1 DennisBA.pdf: 1210954 bytes, checksum: f42a19fb396d47e801ab673ab1f88887 (MD5) Previous issue date: 2005-10-14 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / 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. / Em sistemas multi-rob?s a arquitetura de controle e a estrat?gia de trabalho representam um desafio para os pesquisadores. ? importante que a arquitetura de controle seja robusta, de forma que se adapte naturalmente ?s mudan?as nas caracter?sticas do problema e tamb?m que a estrat?gia de trabalho permita aos rob?s desenvolver as tarefas atribu?das eficaz e eficientemente, levando em considera??o a restri??o de que os rob?s v?o interagir diretamente em ambientes povoados de seres humanos. Neste contexto, este trabalho explora duas abordagens para a coordena??o de uma frota de rob?s desenvolvendo tarefas cooperativas. Ambas as abordagens s?o baseadas em uma mistura de aprendizado por imita??o e por experi?ncia. Assim, na primeira abordagem desenvolveu-se uma arquitetura de controle, uma m?quina de infer?ncia difusa para reconhecimento de fatos em jogos de futebol, um software narrador de jogos baseado na m?quina de infer?ncia difusa, e a implementa??o de aprendizado por imita??o a partir de observa??o e an?lise de outros times rob?ticos. Al?m disso, aplicou-se eficientemente abstra??o de estados em aprendizado por refor?o no problema padr?o de futebol de rob?s. Finalmente, o aprendizado por refor?o foi implementado de forma que as a??es somente s?o executadas em certos estados (por exemplo os estados onde algum sistema rob?tico especialista j? as utilizou) diferentemente da forma tradicional onde as a??es no banco de conhecimento t?m que ser testadas em todos os estados. No caso da segunda abordagem, implementou-se aprendizado por refor?o com aproxima??o de fun??es, para o que foi criado um algoritmo chamado RBF-Sarsa($lambda$). Em ambas as abordagens implementou-se o aprendizado por refor?o em lotes e o aprendizado por imita??o como semente para aprendizado por refor?o. Al?m disso, explorou-se o aprendizado com times de rob?s controlados por seres humanos. As propostas deste trabalho mostraram-se eficientes no problema padr?o de futebol de rob?s, e ao serem implementadas em outros sistemas rob?ticos permitir?o que os mesmos sejam eficazes e eficientes no desenvolvimento das tarefas atribu?das com um alto grau de adapta??o ?s mudan?as dos requerimentos e do ambiente.