Thesis (Ph.D.)--University of Western Australia, 2003.

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1994. / Vita. Abstract. Includes bibliographical references (leaves 388-392). Also available via the Internet.

Thesis (Ph.D)--Aerospace Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Tsiotras, Panagiotis; Committee Member: Corban, Eric; Committee Member: Feron, Eric; Committee Member: Johnson, Eric; Committee Member: Vachtsevanos, George.

Thesis (M.S.)--University of Florida, 2004. / Title from title page of source document. Document formatted into pages; contains 88 pages. Includes vita. Includes bibliographical references.

Thesis (M.S.)--University of Florida, 2004. / Title from title page of source document. Document formatted into pages; contains 42 pages. Includes vita. Includes bibliographical references.

Thesis (M.S.)--University of Florida, 2004. / Title from title page of source document. Document formatted into pages; contains 44 pages. Includes vita. Includes bibliographical references.

Thesis (M.S.)--University of Texas at El Paso, 2009. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Thesis (Ph.D.)--University of Nebraska-Lincoln, 2007. / Title from title screen (site viewed Aug. 2, 2007). PDF text: xviii, 388 p. : ill. UMI publication number: AAT 3258404. Includes bibliographical references. Also available in microfilm and microfiche formats.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1994. / Vita. Abstract. Includes bibliographical references (leaves 107-112). Also available via the Internet.

Made available in DSpace on 2015-03-03T11:52:53Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-07-22Bitstream added on 2015-03-03T12:06:49Z : No. of bitstreams: 1 000802248.pdf: 1455012 bytes, checksum: 1ff27825cc53baa7bf760b7595037fde (MD5) / O Brasil possui um grande potencial no setor do agronegócio e a associação desse setor com o desenvolvimento tecnológico deu origem à Agricultura de Precisão. Nesse contexto, o uso de sensores de Nitrogênio foliar de tempo real, especificamente os que utilizam como princípio de funcionamento e espectrometria, em conjunto com sistemas inteligentes computacionais, tem contribuído de forma decisiva para o incremento da produtividade no campo, evitando a aplicação excessiva de insumos e, assim, preservando o meio ambiente. Um insumo comumente aplicado na cultura de cana-de-açúcar é o Nitrogênio que, apesar de ter grande contribuição econômica, impõe grande impacto ao meio ambiente, principalmente na poluição de aquíferos e mananciais. Dessa maneira, a quantidade aplicada desse nutriente é de grande importância, pois sua falta limita o crescimento da cultura e seu excesso polui o meio ambiente. A determinação da quantidade de Nitrogênio pode ser feita por meio do uso de sensores espectrométricos na faixa do infravermelho próximo visando a cobertura verde da cultura. Entretanto, no estágio inicial de crescimento, a cobertura verde não é plena, de forma que o sensor detecta, além da cobertura verde, o solo e cobertura morta, acrescentando ruído à medida da refletância usada para a estimação do teor de Nitrogênio na planta. Nesse cenário, este trabalho tem o objetivo de mapear a relação entre os valores fornecidos pelo sensor N-SENSOR ALS do fabricante norueguês YARA e os teores reais de Nitrogênio na planta medidos em laboratório. Mais especificamente, sistema de inferência neuro-fuzzy adaptativo (ANFIS), redes neurais artificiais do tipo Perceptron de Múltiplas Camadas (PMC) e General Regression Neural Network (GRNN) serão empregados visando a identificação e o aprendizado da relação entre os valores medidos pelo sensor N-SENSOR ALS e os valores reais obtidos em laboratório, eliminando os ruídos... / Brazil has a great potential in the agribusiness sector and the association of this sector with technological development gave rise to precision agriculture. In this context, the use of soil sensors for real-time, specifically those using operating principle as spectrometry, together with computational intelligent systems are contributing decisively to increasing productivity in the field, avoiding excessive use of inputs and thus preserving the environment. An ingredient commonly used in the cultivation of sugar cane is the nitrogen that, depite great economic contribution, imposes great impact on the environment, especially in aquifers and fountains pollution. Thus, the applied amount of this nutrient is of great importance, since the lack of limits crop growth and excess pollute the environment. The determination of the nitrogen content can be made through the use of spectrometric sensors in the near-infrared aiming the green cover crop. However, in the initial stage of growth the green coverage is not complete, so the sensor sees beyond the green cover, soil and mulch, adding noise to the measurement of reflectance used to estimate the percentage of nitrogen in plant. In this scenario, this paper aims to map the relationship between the values provided by the sensor N-SENSOR ALS Norwegian YARA manufacturer and the actual levels of nitrogen in the plant measured in the laboratory. More specifically, systems of inference adaptive neuro-fuzzy (ANFIS), artificial neural network Multilayer Perceptron type (MLP) and General Regression Neural Network (GRNN) are employed in order to identify learning and the relationship between the values measured by the sensor N-SENSOR ALS and actual obtaind in the laboratory, eliminating the noise imposed by non-green roof and external disturbances such as the variation of ambient light. The results indicate that the neuro-fuzzy approach has superior performance and neural networks can be used to correct...