There is a growing interest in employing sub-soil sensing systems to support precision agriculture. This thesis presents the design of an RFID sub-soil sensing system which is based on integrating passive RFID technology and sub-soil sensing technology. The proposed RFID sub-soil system comprises of an above-ground RFID reader and a number of RFID sub-soil sensor nodes. The key feature of the system is that the sensor nodes do not require an on-board battery, as they are capable of harvesting energy from the ElectroMagnetic (EM) field generated by the RFID reader. The sensor nodes then transmit sensor measurements to the reader wirelessly through soil. With the proposed RFID sub-soil system, the high path loss of the sub-soil wireless channel is a significant problem which leads to the challenge for the system to achieve an acceptable Quality of Service (QoS). In this project, the path loss in soil has been characterised through CST simulations. In the simulations, the effect of the soil on the sensor node antenna has also been investigated. This thesis also presents the design and implementation of a programmable RFID reader platform and an embedded RFID sensor node prototype. The RFID reader platform is implemented using a National Instruments (NI) PXI system, and it is configured and controlled by NI LabVIEW software. The sensor node prototype is capable of harvesting RF energy and transmitting sensor measurements from a temperature sensor through backscatter communication. A series of sub-soil experiments have been carried out to evaluate the performance of the RFID sensor node prototype using the PXI-based RFID reader platform. The experimental results are presented and analysed in this thesis. Additionally, this work has explored trade-offs in the system design, and these design trade-offs are summarised and described.

Protection management with the aid of plant protection products makes it possible to carry out pest control programs in agricultural environments and make them less hazardous for the cultivation of products on a large scale. However, when these programs are put into effect, only a small proportion of the sprayed products is really deposited on the target area while much of it is carried to neighboring regions. The scientific literature includes studies on the use of mathematical techniques to calculate the physical transformation and movement and provide a deposition estimate of the product. On the basis of this prediction, it is possible to configure a system which can allow the spraying to be carried out in normal weather conditions in the region for a satisfactory performance, although these conditions can undergo changes and make any statistical configuration unreliable. An alternative way of overcoming this problem, is to adapt the spray elements to the meteorological conditions while the protection management is being undertaken. However, the current techniques are operationally expensive in computational terms, which makes them unsuitable for situations where a short operational time is required. This thesis can be characterized as descriptive and seeks to allow deposition predictions to be made in a rapid and precise way. Thus it is hoped that the new approaches can enable the spray element to be adapted to the weather conditions while the protection management is being carried out. The study begins by attempting to reduce costs through a computational model of the environment that can speed up its execution. Subsequently, this computational model is used for predicting the rate of deposition as a fitness function in meta-heuristic algorithms and ensure that the mechanical behavior of the spray element can be adapted to the weather conditions while the management is put into effect. The results of this approach show that it can be adapted to environments with low variability. At the same time, it has a poor performance in environments with a high variability of weather conditions. A second approach is investigated and analyzed for this scenario, where the adaptation requires a reduced execution time. In this second approach, a trained machine learning technique is employed together with the results obtained from the first approach in different scenarios. These results show that this approach allows the spray element to be adapted in a way that is compatible with what was provided by the previous approach in less space of time. / O manejo de proteção com uso de produtos fitofarmacêuticos possibilita o controle de pragas em ambientes agrícolas, tornando-o menos nocivo para o desenvolvimento da cultura e com produção em grande escala. Porém, apenas uma pequena parte do produto pulverizado realmente é depositado na área alvo enquanto a maior parte do produto sofre deriva para regiões vizinhas. A literatura científica possui trabalhos com o uso de técnicas matemáticas para calcular a transformação física e movimento para estimar a deposição do produto. Com base nessa predição é possível configurar o sistema de pulverização para realizar a pulverização sob uma condição meteorológica comum na região para um desempenho satisfatório, mas as condições meteorológicas podem sofrer alterações e tornar qualquer configuração estática ineficiente. Uma alternativa para esse problema é realizar a adaptação da atuação do elemento pulverizador às condições meteorológicas durante a execução do manejo de proteção. Contudo, as técnicas existentes são computacionalmente custosas para serem executadas, tornando-as inadequadas para situações em que é requerido baixo tempo de execução. Esta tese se concentra no contexto descrito com objetivo de permitir a predição da deposição de forma rápida e precisa. Assim, espera-se que as novas abordagens sejam capazes de possibilitar a adaptação do elemento pulverizador às condições meteorológicas durante a realização do manejo de proteção. Este trabalho inicia com o processo de redução do custo de execução de um modelo computacional do ambiente, tornando sua execução mais rápida. Posteriormente, utiliza-se este modelo computacional para predição da deposição como função Fitness em algoritmos de meta-heurística para adaptar o comportamento do elemento pulverizador às condições meteorológicas durante a realização do manejo. Os resultados desta abordagem demonstram que é possível utilizá-la para realizar a adaptação em ambientes com baixa variabilidade. Por outro lado, pode apresentar baixo desempenho em ambientes com alta variabilidade nas condições meteorológicas. Uma segunda abordagem é investigada e analisada para este cenário, onde o processo de adaptação requer um tempo de execução reduzido. Nesta segunda abordagem é utilizado uma técnica de Aprendizado de Máquina treinada com os resultados gerados pela primeira abordagem em diferentes cenários. Os resultados obtidos demonstram que essa abordagem possibilita realizar a adaptação do elemento pulverizador compatível com a proporcionada pela abordagem anterior em um menor espaço de tempo.

A aplicação de fertilizantes nitrogenados da forma como é realizada hoje, tem baixa eficiência, causando grandes perdas para o ambiente e aumentando o custo de produção para o agricultor. Novas tecnologias vêm sendo constantemente desenvolvidas para otimizar o manejo de fertilizantes na agricultura. Os sensores ópticos são uma dessas ferramentas que tem potencial para contribuir para a gestão da adubação nitrogenada. Este trabalho consistiu em avaliar o comportamento do NDVI (Normalized Difference Vegetation Index) gerado pelo sensor óptico ativo terrestre Crop Circle® (Holland Scientific, Inc., Lincoln, NE) na cultura do milho. Esse sensor utiliza dois comprimentos de onda centralizados no âmbar (590 ± 5 nm) e no infravermelho próximo (880 ± 10 nm). Ensaios em parcelas foram realizados ao longo de três anos no Campo Demonstrativo e Experimental da Fundação ABC localizado em Itaberá, SP. Os ensaios contemplaram doses de nitrogênio na cultura do milho em dois espaçamentos entre fileiras, 0,80 m e um espaçamento reduzido com 0,40 m, além de outras variáveis avaliadas, que se mostraram importantes no decorrer da execução dos experimentos. A análise estatística dos dados foi feita por regressão (linear, polinomial quadrática e exponencial), correlação e teste de Tukey. Como os ensaios foram realizados em três safras diferentes, um deles aconteceu em uma safra com quantidade de precipitação 53% acima da média histórica, e outro foi realizado em uma gleba com baixo teor de matéria orgânica na camada superficial (0 a 0,2 m) visando obter uma resposta maior à aplicação de nitrogênio. Notou-se que o NDVI tem potencial para uso na cultura do milho, principalmente em anos de baixa resposta às doses de N aplicado. Entretanto o NDVI saturou no ensaio de maior resposta, não identificando os tratamentos com maior dose de nitrogênio devido à grande quantidade de biomassa. Foram encontrados valores de até 75% de economia no uso do nitrogênio ao utilizar a dose correta dos fertilizantes nitrogenados. Entretanto, a alta variação de resultados encontrados entre os três cenários dificulta a geração de um modelo simples para recomendação de nitrogênio em taxa variável. É necessário, portanto, a realização de mais ensaios em uma variação ainda maior de ambientes, levando em consideração outras variáveis com potencial de serem utilizadas em conjunto com o NDVI na geração dos modelos de recomendação, como por exemplo, a matéria orgânica e a disponibilidade hídrica. / Nitrogen application, in the way it is used today has low efficiency, causing great loss to the environment and increasing the production cost to farmers. New technologies have been constantly developed to optimize fertilizers management in agriculture. The optical sensors are one of the tools with potential to contribute for nitrogen management. This work consisted in evaluate the NDVI (Normalized Difference Vegetation Index) from ground active optical sensor Crop Circle® (Holland Scientific, Inc., Lincoln, NE) in corn crops. This sensor uses two wavelengths centered in amber (590 ± 5 nm) and near infrared (880 ± 10 nm). Split-plot experiments were conducted along three years at an experimental field from Fundação ABC, located in Itaberá, São Paulo State, Brazil. Experiments had nitrogen rates in corn, two row spacing, 0.4 and 0.8 m, and other variables were also analyzed. Statistical analyzes were performed by regressions (linear, quadratic and exponential), correlations and Tukey test. As experiments were conducted in three different seasons, one had rainfall 53% over the historic average, and another had low soil organic matter content at (0 to 0.2 m depth) expecting to show higher nitrogen response by the crop. It was noticed that NDVI has potential to be used in corn crops, mainly in low response years to N applied. However, NDVI saturated at the highest response experiment, failing in identify treatments with higher N rates, due the amount of biomass. Nitrogen saving of up to 75% was observed when using the optimum N rate. But the high variations in the results among the three seasons indicate that it is difficult to create a simple algorithm to recommend nitrogen. More experiments in a larger range of environments are necessary, taking in consideration other variables with potential to be used together with NDVI in the generation of recommendations, like organic matter and water availability.

<p>Food security is a major concern of human civilization. A way to ensure food security is to grow plants in a greenhouse under controlled conditions. Even under careful greenhouse production, stress in plants can emerge, and can cause damaging disease. To prevent yield loss farmers, apply resources, e.g., water, fertilizers, pesticides, higher/lower humidity, lighting, and temperature, uniformly in the infected areas. Research, however, shows that the practice leads to non-optimal profit and environmental protection.</p><p>Precision agriculture (PA) is an approach to address such challenges. It aims to apply the right amount or recourses at the right time and place. PA has been able to maximize crop yield while minimizing operation cost and environmental damage. The problem is how to obtain timely, precise information at each location to optimally treat the plants. There is scant research addressing strategies, algorithms, and protocols for analytics in PA. A monitoring and treating systems are the foci of this dissertation.</p><p>The designed systems comprise of agent- and system-level protocols and algorithms. There are four parts: (1) Collaborative Control Protocol for Cyber-Physical System (CCP-CPS); (2) Collaborative Control Protocol for Early Detection of Stress in Plants (CCP-ED); (3) Optimal Inspection Profit for Precision Agriculture; and (4) Multi-Agent System Optimization in Greenhouse for Treating Plants. CCP-CPS, a backbone of the system, establishes communication line among agents. CCP-ED optimizes the local workflow and interactions of agents. Next, the Adaptive Search algorithm, a key algorithm in CCP-ED, has analyzed to obtain the optimal procedure. Lastly, when stressed plants are detected, specific agents are dispatched to treat plants in a particular location with specific treatment. </p><p>Experimental results show that collaboration among agents statistically and significantly improves performance in terms of cost, efficiency, and robustness. CCP-CPS stabilizes system operations and significantly improves both robustness and responsiveness. CCP-ED enabling collaboration among local agents, significantly improves the number of infected plants found, and system efficiency. Also, the optimal Adaptive Search algorithm, which considers system errors and plant characteristics, significantly reduces the operation cost while improving performance. Finally, with collaboration among agents, the system can effectively perform a complex task that requires multiple agents, such as treating stressed plants with a significantly lower operation cost compared to the current practice.</p>

<p> </p> <p>This paper examines the relationship between the Federal Crop Insurance Program (FCIP) participation and technology adoption patterns, using farm-level data from the</p> <p>United States Department of Agriculture (USDA) Agricultural Resource Management Survey (ARMS). Participation in the federally subsidized crop insurance program may be correlated with technology adoption and other various risk management practices. Existing studies indicate that the subsidized FCIP may disincentivize producers from utilizing technology as a risk management tool. Empirical results indicate that producers enrolled in federal crop insurance programs may be more likely to have adopted PATs earlier than producers who were not enrolled in the FCIP. This could indicate that producers may not view the FCIP as a substitute for other risk management options, or that these producers may not view these technologies in the same risk-reducing lens as they may view the FCIP.</p>

From 2019 to 2020, greenhouse and field research was conducted at North Dakota State University to investigate the canopy temperature response of waterhemp (Amaranthus rudis), kochia (Kochia scoparia), common ragweed (Ambrosia artemisiifolia), horseweed (Conyza canadensis), Palmer amaranth (Amaranthus palmeri), and red root pigweed (Amaranthus retroflexus) after glyphosate application to identify glyphosate resistance. In these experiments, thermal images were captured of randomized glyphosate resistant populations and glyphosate susceptible populations of each weed species. The weed canopies' thermal values were extracted and submitted to statistical testing and various classifiers in an attempt to discriminate between resistant and susceptible populations. Glyphosate resistant horseweed, when collected within greenhouse conditions, was the only biotype reliably classified using significantly cooler temperature signatures than its susceptible counterpart. For field conditions, image based machine learning classifiers using thermal data were outperformed by classifiers made using additional multispectral data, suggesting thermal is not a reliable predictor of glyphosate resistance.

京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第19771号 / 農博第2167号 / 新制||農||1040(附属図書館) / 学位論文||H28||N4987(農学部図書室) / 32807 / 京都大学大学院農学研究科地域環境科学専攻 / (主査)教授 飯田 訓久, 教授 近藤 直, 准教授 中村 公人 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM

With an increased interest in precision agriculture, it is important to identify efficient ways to monitor soil moisture. Soil moisture can be monitored using handheld sensors, but this method is laborious and time consuming. Remote methods, such as radar systems can be used as well, but these methods require ground truth data to verify their accuracy. It becomes clear that to collect this data regularly and reliably, a mobile robotic device is necessary. This thesis proposes to implement mobile robot take soil moisture measurements with less human effort than existing methods while maintaining the same accuracy. This soil moisture data collection system uses an unmanned ground vehicle (UGV) to take measurements with position data. This system uses an actuator inserted soil moisture probe, and a radio frequency identification (RFID) sensing system that uses buried moisture sensing tags. Field testing of both measurement systems showed that the actuator-based system worked reliably.

Precision agriculture continues to be prevalent within row-crop production. The purpose of this study was to investigate the adoption status of precision agricultural practices among selected row-crop (soybean, wheat, corn, cotton, peanuts, and rice) producers in Alabama, Arkansas, Georgia, Louisiana, Mississippi, and Tennessee. Seventy-four percent of row-crop producers surveyed in this study had adopted precision agricultural practices in their farming operations. Eighty-three percent of respondents indicated they were using automated GPS technology such as autosteer, 66% were using manual guidance such as lightbar, 63% variable-rate prescription map, and 34% auto-sprayer boom section or nozzle control. The primary source for receiving information relating to precision agriculture were agricultural dealerships, extension, and crop consultants, respectively. In addition, the amount of acreage a producer farmed was a statistically significant predictor of how many precision agricultural technologies a producer adopted.

United States Department of Agriculture’s Farm Bill conservation programs provide landowner incentives to remove less productive and environmentally sensitive lands from agricultural production and re-establish them in natural vegetation to achieve conservation objectives. However, removal of arable land from production imposes an opportunity cost associated with loss in revenue from commodities that otherwise would have been produced. The Habitat Buffers for Upland Birds practice (CP-33) under the Continuous Conservation Reserve Program is a targeted conservation practice designed to increase northern bobwhite populations in agricultural landscapes. However, establishing CP-33 buffers on profitable farmland may be incompatible with economic objectives of landowners. To determine how CP-33 enrollment influenced field profitability and bobwhite abundance; I simulated CP-33 buffers on crop fields across a range of commodity prices and modeled profitability and predicted bobwhite abundance. CP-33 increased field revenue on a percentage of fields at all commodity prices and increased bobwhite abundance up to 30%.