As técnicas utilizadas na agricultura de precisão agregam não só ferramentas para o diagnóstico das causas da variabilidade encontrada nas lavouras, como também soluções para o convívio ou correção dessa variabilidade. O Brasil é o maior produtor mundial de café e ainda hoje apresenta um mercado em franca expansão. Por isso, em razão do potencial produtivo e da lucratividade da cultura do café, atualmente o interesse pelas técnicas de manejo localizado e a procura por novas tecnologias estão aumentando. O objetivo do presente trabalho foi avaliar a influência do manejo localizado das adubações de nitrogênio, fósforo e potássio, na produtividade e no estado nutricional do cafeeiro. Para isso, foram utilizadas quatro áreas comerciais, sendo uma localizada no município de Gália, SP, conduzida durante quatro safras, e três no município de Patrocínio, MG, conduzidas durante três safras. Estabeleceram-se dois tratamentos para testar a hipótese do aumento da produtividade e a manutenção do estado nutricional do cafeeiro com aplicações localizadas dos nutrientes nitrogênio, fósforo e potássio, em doses variáveis. Como resultados foram observados que as adubações de nitrogênio, fósforo e potássio aplicadas em doses variáveis aumentaram a produtividade do cafeeiro em 240,0 kg ha-1, quando somadas todas as safras. Houve redução do consumo do nitrogênio em 134,7 kg ha-1 e do potássio em 82,0 kg ha-1, e para o nutriente fósforo ocorreu um aumento no consumo de 65,0 kg ha-1. A avaliação do estado nutricional das plantas de cafeeiro mostrou-se apropriada para possíveis correções das doses dos nutrientes durante a adubação. / The precision agriculture techniques, besides adding tools to identify the causes of variability found in the fields, provide solutions for this variability to be corrected or for the best management solution when the variability is unavoidable. Brazil is the biggest producer of coffee within an increasing market. Therefore, because of the potential yield and profit of coffee crop, currently the interest in site specific management techniques and the demand for new technologies are increasing. This study aimed to evaluate the influence of site specific management of fertilizers (nitrogen, phosphorus and potassium) in the yield and in the nutritional status of coffee crop. Four commercial areas were used, one located in Gália, SP, and three in Patrocínio, MG. Two treatments were established to test the hypothesis of higher yield and maintenance of the nutritional status of coffee with site specific fertilizer rates of nitrogen, phosphorus and potassium. The results indicated that nitrogen, phosphorus and potassium nutrients applied in variable rates increased the yield in 240.0 kg ha-1, considering all the harvesting seasons. Nitrogen fertilizer consumption was reduced in 134.7 kg ha-1 and potassium in 82.0 kg ha-1, on the other hand phosphorus consumption increased in 65.0 kg ha-1. The evaluation of nutritional status of coffee plants was appropriate for possible corrections of nutrient supplying during fertilization.

Irrigation needs to be applied to soils in relatively humid regions such as western Kentucky to supply water for crop uptake to optimize and stabilize yields. Characterization of soil and crop variability at the field scale is needed to apply site specific management and to optimize water application. The objective of this work is to propose a characterization and modeling of soil and crop processes to improve irrigation management. Through an analysis of spatial and temporal behavior of soil and crop variables the variability in the field was identified. Integrative analysis of soil, crop, proximal and remote sensing data was utilized. A set of direct and indirect measurements that included soil texture, electrical conductivity (EC), soil chemical properties (pH, organic matter, N, P, K, Ca, Mg and Zn), NDVI, topographic variables, were measured in a silty loam soil near Princeton, Kentucky. Maps of measured properties were developed using kriging, and cokriging. Different approaches and two cluster methods (FANNY and CLARA) with selected variables were applied to identify management zones. Optimal scenarios were achieved with dividing the entire field into 2 or 3 areas. Spatial variability in the field is strongly influenced by topography and clay content. Using Root Zone Water Quality Model 2.0 (RZWQM), soil water tension was modeled and predicted at different zones based on the previous delineated zones. Soil water tension was measured at three depths (20, 40 and 60 cm) during different seasons (20016 and 2017) under wheat and corn. Temporal variations in soil water were driven mainly by precipitation but the behavior is different among management zones. The zone with higher clay content tends to dry out faster between rainfall events and reveals higher fluctuations in water tension even at greater depth. The other zones are more stable at the lower depth and share more similarities in their cyclic patterns. The model predictions were satisfactory in the surface layer but the accuracy decreased in deeper layers. A study of clay mineralogy was performed to explore field spatial differences based on the map classification. kaolinite, vermiculite, HIV and smectite are among the identified minerals. The clayey area presents higher quantity of some of the clay minerals. All these results show the ability to identify and characterize the field spatial variability, combining easily obtainable data under realistic farm conditions. This information can be utilized to manage resources more effectively through site specific application.

Herbicides applications could possibly be reduced if targeted. Targeting the applications requires prior identification and quantification of the weed population. This task could possibly be done by a weed scout robot. The ability to position a camera over the inter-row space of densely seeded crops will help to simplify the task of automatically quantifying weed infestations. As part of the development of an autonomous weed scout, a vision-based local positioning system for weed detection has been developed and tested in a laboratory setting. Four Line-detection algorithms have been tested and a robotic positioning device, or XYZtheta-table, was developed and tested. <p> The Line-detection algorithms were based respectively on a stripe analysis, a blob analysis, a linear regression and the Hough Transform. The last two also included an edge-detection step. Images of parallel line patterns representing crop rows were collected at different angles, with and without weed-simulating noise. The images were processed by the four programs. The ability of the programs to determine the angle of the rows and the location of an inter-row space centreline was evaluated in a laboratory setting. All algorithms behaved approximately the same when determining the rows angle in the noise-free images, with a mean error of 0.5°. In the same situation, all algorithms could find the centreline of an inter-row space within 2.7 mm. Generally, the mean errors increased when noise was added to the images, up to 1.1° and 8.5 mm for the Linear Regression algorithm. Specific dispersions of the weeds were identified as possible causes of increase of the error in noisy images. Because of its insensitivity to noise, the Stripe Analysis algorithm was considered the best overall. The fastest program was the Blob Analysis algorithm with a mean processing time of 0.35 s per image. Future work involves evaluation of the Line-detection algorithms with field images. <p>The XYZtheta-table consisted of rails allowing movement of a camera in the 3 orthogonal directions and of a rotational table that could rotate the camera about a vertical axis. The ability of the XYZtheta-table to accurately move the camera within the XY-space and rotate it at a desired angle was evaluated in a laboratory setting. The XYZtheta-table was able to move the camera within 7 mm of a target and to rotate it with a mean error of 0.07°. The positioning accuracy could be improved by simple mechanical modifications on the XYZtheta-table.

Labor shortage has prompted researchers to develop robot platforms for agriculture field scouting tasks. Sensor-based automatic topographic mapping and scouting algorithms for rough and large unstructured environments were presented. It involves moving an image sensor to collect terrain and other information and concomitantly construct a terrain map in the working field. In this work, a triangular mesh map was first used to represent the rough field surface and plan exploring strategies. A 3D image sensor model was used to simulate collection of field elevation information.<p>A two-stage exploring policy was used to plan the next best viewpoint by considering both the distance and elevation change in the cost function. A greedy exploration algorithm based on the energy cost function was developed; the energy cost function not only considers the traveling distance, but also includes energy required to change elevation and the rolling resistance of the terrain. An information-based exploration policy was developed to choose the next best viewpoint to maximise the information gain and minimize the energy consumption. In a partially known environment, the information gain was estimated by applying the ray tracing algorithm. The two-part scouting algorithm was developed to address the field sampling problem; the coverage algorithm identifies a reasonable coverage path to traverse sampling points, while the dynamic path planning algorithm determines an optimal path between two adjacent sampling points.<p>The developed algorithms were validated in two agricultural fields and three virtual fields by simulation. Greedy exploration policy, based on energy consumption outperformed other pattern methods in energy, time, and travel distance in the first 80% of the exploration task. The exploration strategy, which incorporated the energy consumption and the information gain with a ray tracing algorithm using a coarse map, showed an advantage over other policies in terms of the total energy consumption and the path length by at least 6%. For scouting algorithms, line sweeping methods require less energy and a shorter distance than the potential function method.

Labor shortage has prompted researchers to develop robot platforms for agriculture field scouting tasks. Sensor-based automatic topographic mapping and scouting algorithms for rough and large unstructured environments were presented. It involves moving an image sensor to collect terrain and other information and concomitantly construct a terrain map in the working field. In this work, a triangular mesh map was first used to represent the rough field surface and plan exploring strategies. A 3D image sensor model was used to simulate collection of field elevation information.<p>A two-stage exploring policy was used to plan the next best viewpoint by considering both the distance and elevation change in the cost function. A greedy exploration algorithm based on the energy cost function was developed; the energy cost function not only considers the traveling distance, but also includes energy required to change elevation and the rolling resistance of the terrain. An information-based exploration policy was developed to choose the next best viewpoint to maximise the information gain and minimize the energy consumption. In a partially known environment, the information gain was estimated by applying the ray tracing algorithm. The two-part scouting algorithm was developed to address the field sampling problem; the coverage algorithm identifies a reasonable coverage path to traverse sampling points, while the dynamic path planning algorithm determines an optimal path between two adjacent sampling points.<p>The developed algorithms were validated in two agricultural fields and three virtual fields by simulation. Greedy exploration policy, based on energy consumption outperformed other pattern methods in energy, time, and travel distance in the first 80% of the exploration task. The exploration strategy, which incorporated the energy consumption and the information gain with a ray tracing algorithm using a coarse map, showed an advantage over other policies in terms of the total energy consumption and the path length by at least 6%. For scouting algorithms, line sweeping methods require less energy and a shorter distance than the potential function method.

A method for precisely synchronizing an external serial data stream to the pulse-per-second (PPS) output signal from a global navigation satellite-based system (GNSS) receiver was investigated. A signal timing device was designed that used a digital signal processor (DSP) with serial inputs and input captures to generate time stamps for asynchronous serial data based on an 58593.75 Hz internal timer. All temporal measurements were made directly in hardware to eliminate software latency. The resolution of the system was 17.1 µs, which translated to less than one millimeter of horizontal position error at travel speeds typical of most agricultural operations. The dynamic error of a TTS was determined using a rotary test fixture. Tests were performed at angular velocities ranging from 0 to 3.72 rad/s and a radius of 0.635 m. Average latency from the TTS was shown to be consistently near 0.252 s for all angular velocities and less variable when using a reflector based machine target versus a prism target. Sight distance from the target to the TTS was shown to have very little effect on accuracy between 4 and 30 m. The TTS was determined to be a limited as a position reference for dynamic GNSS and vehicle auto-guidance testing based on angular velocity. The dynamic error of a GNSS receiver was determined using the rotary test fixture and modeled as discrete probability density functions for varying angular velocities and filter levels. GNSS position and fixture data were recorded for angular velocities of 0.824, 1.423, 2.018, 2.618, and 3.222 rad/s at a 1 m radius. Filter levels were adjusted to four available settings including; no filter, normal filter, high filter, and max filter. Each data set contained 4 hours of continuous operation and was replicated three times. Results showed that higher angular velocities increased the variability of the distribution of error while not having a significant effect on average error. The distribution of error tended to change from normal distributions at lower angular velocities to uniform distributions at higher angular velocities. Internal filtering was shown to consistently increase dynamic error for all angular velocities.

The Precision Agriculture is, within the Agribusiness, one of the drivers of technological development and sustainable production. It is characterized as a set of technologies applied in a farm management system, which allows the results to be the most significant potential in a framework of possibilities where solving strategies of uniformity problems of crops can be addressed. This knowledge of the variability of production and quality, which consists in the use of machinery and equipment, is useful for any culture in small or large scale. The Precision Agriculture is not only about the use of sophisticated machinery and equipment, but also management and administration of the Agricultural Production Unit. With special attention to the need of approximate the knowledge and improvement in precision agriculture, the main objective of this work was offer a new school subject called "Management of Precision Agriculture". The class was offered in the first term of 2016, within the curriculum for the Technical Course in Agriculture in the State Technical School Nossa Senhora da Conceição , Três Vendas district of Cachoeira do Sul RS. The subject, with a workload of 40 hours was planned based on theoretical and practical notions in: Introduction and AP Cycle, Global Navigation Satellite Systems (GNSS), the Aquarius Project History, Precision Agriculture applied in soil science, GIS, Basic Statistics, Agriculture in Machinery and Tools precision, Informatics applied to Precision Agriculture, Information Management, agricultural Equipment Maintenance precision. Evaluation methods of the subject were the application of essay tests, presentation of seminars, preparation of summaries of important topics and also practical classes. As main results, we highlight the good acceptance of the subject for students and people involved in the process, 100% approval of the students and parameters to improve the subject, including the possibility of offering it in the next class. / A Agricultura de Precisão constitui dentro do Agronegócio, uma das forças propulsoras do desenvolvimento tecnológico e sustentável da produção. Caracteriza-se como um conjunto de tecnologias aplicadas em um sistema de gerenciamento agrícola, que permite que os resultados sejam os mais significativos possíveis em um quadro de possibilidades onde estratégias de resolução dos problemas da desuniformidade das lavouras possam ser equacionados. Esse conhecimento da variabilidade da produção e da qualidade consiste no uso de máquinas e equipamentos útil para qualquer cultura em pequena e ou em grande escala. Porém, a Agricultura de Precisão não é apenas o uso de máquinas e equipamentos sofisticados, mas também gestão e administração da Unidade de Produção Agrícola. Destacando-se a necessidade de tornar mais próximo o conhecimento e o aperfeiçoamento em Agricultura de precisão. Como objetivo principal desse trabalho, elaborou-se e ofertou-se uma nova disciplina para o Curso Técnico em Agropecuária, já no primeiro semestre de 2016, na Escola Estadual Técnica Nossa Senhora da Conceição, localizada no distrito de Três Vendas, Cachoeira do Sul RS. Elaborou-se e disponibilizou-se a disciplina Gestão em Agricultura de Precisão , com carga horária de 40 horas, trabalhando noções de teoria e prática. A disciplina ofereceu como conteúdo programático: Introdução e Ciclo da AP, Sistemas de Satélites de Navegação Global (GNSS), Histórico do Projeto Aquarius, Agricultura de Precisão aplicada na ciência do solo, Geoprocessamento, Estatística básica, Máquinas e Ferramentas em Agricultura de Precisão, Informática aplicada à Agricultura de precisão, Gestão da Informação, Manutenção de Equipamentos de Agricultura de precisão. Os métodos de análise de resultados da disciplina foram a aplicação de provas descritivas, apresentação de seminários, elaboração de resumos sobre tópicos da disciplina e também aulas práticas. Dentre os resultados, destaca-se ótima aceitação da disciplina pelos estudantes, coordenação do curso e também coordenadores da escola, aprovação de todos os alunos matriculados e uma avaliação para possíveis melhorias futuras da disciplina e possibilidade de oferta da mesma em turmas próximas.

Agribusiness as a whole has evolved along with the global modernization. Precision Agriculture (PA) is a technology that has been adding in yield of major crops such as soybeans, corn, wheat and coffee. The generation of yield maps and maps of fertility are likely to generate information to make further interventions or remedial applications in crops. Aiming to investigate the relationship between yield maps of soybean with chemical indicators of soil fertility work was undertaken on the property of Seeds Bee the municipality of Coxilha - RS. The study began in 2005 until 2013 in the collection of data on productivity done through maps worked with the aid of Ag Leader SMS software, totaling 5 maps of soybean. The yield maps are generated from information collected in the harvest product. These figures are achieved with devices equipped with GPS, humidity sensor and productivity throughout the harvest generate localized information. From this, the maps was unified into one with three classifications of production zones: high, medium and low. With information from areas of high and low production soil samples were collected up to 1 m deep by 5 points geo-referenced each distinct zone (high and low). In the evaluations of the analysis it was realized that the pH and base saturation and sulfur were the attributes that have higher productivity in relation to depth. The match had a positive influence on productivity up to 0.20 m depth. Organic matter showed no similarities with productivity. In view of this correlation with the factors that influence the production allow the producer a decision which product to apply and how much more needs to be based on georeferenced data from your area. This achieves - rational use of natural resources and therefore the environment can still raise productive crops ceilings. / O Agronegócio como um todo vem evoluindo junto com a modernização global. A agricultura de Precisão (AP) é uma tecnologia que vem agregando na produção de grãos das grandes culturas como soja, milho, trigo e café. A geração de mapas de produtividade e de mapas de fertilidade são possibilidades de gerar informações para posteriormente fazer as intervenções ou aplicações de corretivos nas lavouras. Com o objetivo de Investigar a relação entre mapas de produtividade das culturas de soja com indicadores químicos da fertilidade do solo foi realizado um trabalho na propriedade da Sementes Bee no município de Coxilha RS. O estudo iniciou-se em 2005 até 2013 na coleta de informações de produtividade feito através de mapas trabalhados com auxílio do software SMS Ag Leader, totalizando 5 mapas da cultura da soja. Os mapas de produtividade são gerados a partir de informações coletadas na colheita do produto. Esses dados são conseguidos com equipamentos equipados com GPS, sensor de umidade e de produtividade que no decorrer da colheita geram informações localizadas. A partir disso, unificou-se os mapas em um só com 3 classificações de zonas de produção: alta, média e baixa. Com a informação das zonas de alta e baixa produção foram coletadas amostras de solo até 1 m de profundidade em 5 pontos georrefenciados de cada zona distinta (alta e baixa). Nas avaliações das análises percebeu-se que O pH e saturação de bases e enxofre foram os atributos que deram maior relação com a produtividade em profundidade. O fósforo teve influência positiva com a produtividade até os 0,20 m de profundidade. A matéria orgânica não apresentou semelhanças com a produtividade. Em vista disso a correlação com fatores que influenciam na produção possibilitam ao produtor uma decisão de qual produto aplicar e em que quantidade, mais precisa por estar baseada em dados georreferenciados de sua área. Com isso consegue-se racionalizar o uso de recursos naturais e conseqüentemente do meio ambiente podendo ainda elevar os tetos produtivos das lavouras.

OLIVEIRA, Medida de atributos físicos do solo por um sensor capacitivo e aplicação no manejo da irrigação da melancia. 2012. 113 f. : Tese (doutorado) - Universidade Federal do Ceará, Centro de Ciências Agrárias, Departamento de Engenharia Agrícola, Programa de Pós-Graduação em Engenharia Agrícola, Fortaleza-CE, 2012. / Submitted by demia Maia (demiamlm@gmail.com) on 2016-08-02T14:47:59Z No. of bitstreams: 1 2012_tese_adsoliveira.pdf: 9954665 bytes, checksum: 4cf53911bfca35d5e5215682663f8cb9 (MD5) / Approved for entry into archive by demia Maia (demiamlm@gmail.com) on 2016-08-02T14:54:40Z (GMT) No. of bitstreams: 1 2012_tese_adsoliveira.pdf: 9954665 bytes, checksum: 4cf53911bfca35d5e5215682663f8cb9 (MD5) / Made available in DSpace on 2016-08-02T14:54:40Z (GMT). No. of bitstreams: 1 2012_tese_adsoliveira.pdf: 9954665 bytes, checksum: 4cf53911bfca35d5e5215682663f8cb9 (MD5) Previous issue date: 2012 / In recent years, with the explosive development of information technology and electronics, the scientific community worldwide has intensified research to develop and implement research-based improvement in precision application. Advances in information technology, especially in the developing of interfaces with electronic sensors, allowed for the development and application of various transducers in Agricultural Engineering. With respect to irrigated farms there is a major concern with the large volume of water used. According to FAO by 2030 it will be required an increase of 14% of drinking water used by irrigation, in order to meet the demand caused by population growth worldwide. In order to improve water use in irrigation, several studies have been and are being developed, especially focused on evaluating, economically, the conversion of water into food and the actual requirement of crops for their full development. The kc is a factor that relates crop water requirement of a real crop to the water requirement of a hypothetical culture. The soil water balance accounts for all additions and withdrawals of water that actually occur in a certain volume of soil used in agricultural production, allowing to access the water situation in which a culture is. From the above, this study aimed to evaluate the use of a capacitive FDR sensor to replace tensiometer in the process of obtaining the crop coefficient (kc) through the soil water balance. The experiment was conducted at the Low Acaraú Irrigation District - Ceará, in an area of 1.0 ha, into two cycles with of watermelon, in the years 2009 and 2010. The sensors were calibrated in the laboratory using regression analysis between soil moisture from a soil sample and frequency produced by the sensor, and between it and the tension found in a mercury based tensiometer, from saturation to approximately 8.0% volumetric water content. The evaluation of the calibration equation occurred by comparing the sensor data and data from tensiometers for moisture and irrigation timing. The calibration equation for both moisture and for the tension was exponential. The comparison between the estimated and measured tensions using the fitted equation produced a strong performance and/or great on all items analyzed. Field validation of the fitted models was conducted by comparative analysis between the frequencies obtained from the FDR sensors and the tensiometers during the first harvesting year. For the second harvesting year, the technique of filter paper was tested. It resulted in the following equations to fit the water retention curves for the soil: θ=0,3843*T^(-0,381) (1st layer), θ=0,4381*T^(-0,372) (2nd layer) and θ=0,4103*T^(-0,34) (3rd layer). In addition, the instantaneous profile method for evaluating the hydraulic conductivity was applied. Were application depth was applied as a treatment for analyzing some physiological characteristics of the crop. The treatments showed statistical differences for stomatal conductance and transpiration, T4 (ϴ= 1.3 of field capacity) being superior to the others to a 0.05 confidence level. kc values ranged between treatments from 0.74 to 1.06 (18 DAP), 0.98 to 1.52 (45 DAP) and 0.67 to 0.85 (59 DAP). Crop yield ranged from 13.019 to 25.867 Mg between treatments. The efficiency of water use ranged from 4.17 to 6.95 kg m-3. The use of the sensor allows for real time measurement of soil water potential and soil moisture, enabling a more secure monitoring and decision making. / Nos últimos anos, com a evolução explosiva da informática e da eletrônica, a comunidade científica mundial tem intensificado os estudos no sentido de desenvolver e aplicar pesquisas baseada no aprimoramento da alta precisão. Os avanços na área de informática, principalmente na elaboração de interfaces com sensores eletrônicos, permitiram o desenvolvimento e aplicação de vários transdutores na Engenharia Agrícola. No que diz respeito à agricultura irrigada uma grande preocupação é com o enorme volume de água utilizado. Segundo a FAO até 2030 será necessário um incremento de 14% da água potável, na irrigação, a fim de suprir a demanda provocada pelo crescimento da população mundial. No intuito de aprimorar o uso da água na irrigação, vários trabalhos foram e estão sendo desenvolvidos, principalmente na busca de se avaliar, economicamente, a conversão da água em alimento e a real necessidade hídrica das culturas para seu desenvolvimento pleno. O kc é um fator que relaciona a necessidade hídrica da cultura em estudo e a necessidade hídrica de uma cultura hipotética. O balanço hídrico do solo contabiliza todas as adições e retiradas de água que realmente ocorrem em certo volume de solo, utilizado na produção agrícola, permitindo avaliar a situação hídrica pela qual uma cultura passa. Pelo exposto acima, o presente trabalho teve como objetivo geral a avaliação do uso do sensor capacitivo FDR em substituição ao tensiômetro para o cálculo do coeficiente de cultivo (kc), por meio do balanço hídrico do solo. O experimento foi realizado no Perímetro Irrigado do Baixo Acaraú – Ceará, em uma área de 1,0 ha, em dois ciclos com a cultura da melancia nos anos de 2009 e 2010. Os sensores foram calibrados, em laboratório, através de análise de regressão entre umidade da amostra e frequência produzida pelo sensor, e entre esta última e a tensão verificada em um tensiômetro com mercúrio partindo do ponto de saturação até aproximadamente 8,0% de umidade volumétrica. A avaliação da equação de calibração ocorreu através da comparação entre os dados dos sensores e os dados dos tensiômetros para umidade e tempo de irrigação. A equação de calibração tanto para a umidade, como para tensão foi do tipo exponencial. A comparação entre as tensões medidas e estimadas, pela equação de ajuste, alcançou desempenho forte e/ou ótimo em todos os itens analisados. A validação em campo, das equações de ajuste, deu-se através de uma análise comparativa entre as tensões obtidas pelos sensores FDR e pelos tensiômetros, além das umidades atuais, estimadas por ambos os equipamentos, no primeiro ano de cultivo. Para o segundo ciclo produtivo fez-se uso da técnica do papel filtro que retornou as seguintes equações de ajuste para as curvas de retenção da água no solo: θ=0,3843*T^(-0,381)(1ª camada), θ=0,4381*T^(-0,372)(2ª camada) e θ=0,4103*T^(-0,34)(3ª camada), e a técnica do perfil instantâneo para avaliação da condutividade hidráulica. Aplicaram-se variações de umidade do solo, com tratamento, a fim de verificar algumas respostas fisiológicas da cultura. Os tratamentos apresentaram diferença estatística para transpiração e condutância estomática, sendo o T4 (ϴ= 1,3 da capacidade de campo) superior aos demais ao nível de confiança de 0,05. Os valores de kc variaram entre os tratamentos de 0,74 a 1,06 (18º DAP), 0,98 a 1,52 (45º DAP) e 0,67 a 0,85 (59º DAP). A produção variou de 13,019 a 25,867 Mg entre os tratamentos. O uso eficiente de água variou de 4,17 a 6,95 kg m-3. O uso do sensor permite medir o potencial da água no solo e a umidade em tempo real, possibilitando o monitoramento e a tomada de decisões de forma mais segura.

A computação visual é uma área do conhecimento que estuda o desenvolvimento de sistemas artificiais capazes de detectar e desenvolver a percepção do meio ambiente através de informações de imagem ou dados multidimensionais. A percepção visual e a manipulação são combinadas em sistemas robóticos através de duas etapas "olhar"e depois "movimentar-se", gerando um laço de controle de feedback visual. Neste contexto, existe um interesse crescimente no uso dessas técnicas em veículos aéreos não tripulados (VANTs), também conhecidos como drones. Essas técnicas são aplicadas para posicionar o drone em modo de vôo autônomo, ou para realizar a detecção de regiões para vigilância aérea ou pontos de interesse. Os sistemas de computação visual geralmente tomam três passos em sua operação, que são: aquisição de dados em forma numérica, processamento de dados e análise de dados. A etapa de aquisição de dados é geralmente realizada por câmeras e sensores de proximidade. Após a aquisição de dados, o computador embarcado realiza o processamento de dados executando algoritmos com técnicas de medição (variáveis, índice e coeficientes), detecção (padrões, objetos ou áreas) ou monitoramento (pessoas, veículos ou animais). Os dados processados são analisados e convertidos em comandos de decisão para o controle para o sistema robótico autônomo Visando realizar a integração dos sistemas de computação visual com as diferentes plataformas de VANTs, este trabalho propõe o desenvolvimento de um framework para controle de missão e guiamento de VANTs baseado em visão computacional. O framework é responsável por gerenciar, codificar, decodificar e interpretar comandos trocados entre as controladoras de voo e os algoritmos de computação visual. Como estudo de caso, foram desenvolvidos dois algoritmos destinados à aplicação em agricultura de precisão. O primeiro algoritmo realiza o cálculo de um coeficiente de reflectância visando a aplicação auto-regulada e eficiente de agroquímicos, e o segundo realiza a identificação das linhas de plantas para realizar o guiamento dos VANTs sobre a plantação. O desempenho do framework e dos algoritmos propostos foi avaliado e comparado com o estado da arte, obtendo resultados satisfatórios na implementação no hardware embarcado. / Cumputer Vision is an area of knowledge that studies the development of artificial systems capable of detecting and developing the perception of the environment through image information or multidimensional data. Nowadays, vision systems are widely integrated into robotic systems. Visual perception and manipulation are combined in two steps "look" and then "move", generating a visual feedback control loop. In this context, there is a growing interest in using computer vision techniques in unmanned aerial vehicles (UAVs), also known as drones. These techniques are applied to position the drone in autonomous flight mode, or to perform the detection of regions for aerial surveillance or points of interest. Computer vision systems generally take three steps to the operation, which are: data acquisition in numerical form, data processing and data analysis. The data acquisition step is usually performed by cameras or proximity sensors. After data acquisition, the embedded computer performs data processing by performing algorithms with measurement techniques (variables, index and coefficients), detection (patterns, objects or area) or monitoring (people, vehicles or animals). The resulting processed data is analyzed and then converted into decision commands that serve as control inputs for the autonomous robotic system In order to integrate the visual computing systems with the different UAVs platforms, this work proposes the development of a framework for mission control and guidance of UAVs based on computer vision. The framework is responsible for managing, encoding, decoding, and interpreting commands exchanged between flight controllers and visual computing algorithms. As a case study, two algorithms were developed to provide autonomy to UAVs intended for application in precision agriculture. The first algorithm performs the calculation of a reflectance coefficient used to perform the punctual, self-regulated and efficient application of agrochemicals. The second algorithm performs the identification of crop lines to perform the guidance of the UAVs on the plantation. The performance of the proposed framework and proposed algorithms was evaluated and compared with the state of the art, obtaining satisfactory results in the implementation of embedded hardware.