Dissertations / Theses on the topic 'Greenhouse automation'

Expensive and complex automated systems for greenhouses are frequently utilized in the horticulture industry. In parallel, smart systems for home automation has recently seen a rapid increase in popularity. This project aims to combine the climate optimization capabilities of industrial systems with the convenience of home automation systems. More specifically, this project is focused on the design and implementation of electrical and mechanical requirements of a smart greenhouse system. This involved the selection of hardware components, such as sensors, actuators and controllers. It also involved the interconnection between these components and the development of measurement and control systems to autonomously manage the greenhouse. The system is based on a WiFi-connected microcontroller. Parameters monitored include; temperature, humidity and wind-speed. Irrigation is controlled by a solenoid valve and can be scheduled to desired intervals. Ventilation and temperature optimization is done by controlling the roof-hatch with a linear actuator and control of a heater. The results demonstrate a accurate and reliable system with low power consumption. The resulting prototype can be installed in new and existing greenhouses. Functionalities can be remotely controlled and monitored by the convenience of an android application. The total cost of the components used was around 4 500 SEK. Further development could be done to enable seamless scalability with additional components and functionalities. Climate optimization by incorporation of weather forecast as a parameter could be implemented to further reduce energy consumption.
Dyra och komplexa automatiserade växthussystem är vanligt förekommande inom industrin för hortikultur. Parallellt har populariteten för "Smart Home" system för hemautomatisering radikalt ökat. Målet med detta projekt är att kombinera klimatoptimiseringsmöjligheterna hos industriella system med lättanvändligheten hos system för hemautomatisering. Detta projekt fokuserar specifikt på designen och implementationen av de elektriska och mekaniska krav som ställs på ett "smart greenhouse system". Detta innefattar; val av komponenter såsom sensorer, aktuatorer samt styrenhet men även sammankopplingen mellan dessa komponenter och utvecklingen av mjukvara till reglersystemet, som i sin tur syftar till automatiseringen i växthuset. Systemet är baserat på en WiFi-uppkopplad mikrokontroller. Parametrar som monitoreras är; temperatur, luftfuktighet och vindhastighet. Bevattning kontrolleras av en magnetventil och kan schemaläggas för bevattning i önskade intervall. Ventilering och temperaturoptimisering sker genom kontroll av taklucka med hjälp av ett linjärt ställdon samt kontroll av ett värmeelement. Resultatet demonstrerar ett pålitligt och punktligt system med låg energiförbrukning. Prototypen som utvecklats kan installeras i både nya och befintliga växthus. Funktionaliteterna kan smidigt fjärrkontrolleras och monitoreras från en android applikation. Den totala kostnaden för de komponenter som använts var runt 4500 kr. Vidareutveckling vad gäller skalbarhet för att sömlöst lägga till komponenter och funktionaliteter bör övervägas. För ytterligare minskad energiförbrukning med hjälp av klimatoptimering kan väderprognos adderas som en parameter.

A agricultura tradicional realizada em campo aberto é dependente do meio físico natural, sendo sua prosperidade resultado de circunstâncias favoráveis do solo, do clima e água, entre outros. A necessidade crescente de se produzir vegetais com alta qualidade, do planejamento da produção agrícola em termos de quantidade e prazo, da redução dos custos por unidade de produção, com a manutenção ou aumento da qualidade têm levado a um aumento da utilização de cultivo protegido. A realização da produção agrícola com uma certa independência das condições climáticas pode ser obtida através da utilização de casas de vegetação, comercialmente conhecidas como estufas. Diversos estudos mostram que as principais variáveis climáticas envolvidas no processo de produção vegetal são: a temperatura, a umidade, a luminosidade e a concentração de gás carbônico. A presença desses fatores, dentro de certos limites mínimos e máximos, proporciona condições propícias para o desenvolvimento vegetal, enquanto que fora desses limites, o desenvolvimento é prejudicado. Portanto, uma boa política de controle dessas variáveis torna-se imprescindível. Este trabalho apresenta um sistema desenvolvido para a pesquisa em controle climático em casas de vegetação, denominado Agrilogic. Ele utiliza elementos comumente encontrados em automação industrial, como CLP (Controlador Lógico Programável) para as atividades consideradas de curto prazo e um software supervisório do tipo SCADA (Supervisory Control And Data Acquisition) para controlar as atividades consideradas de longo prazo e para a IHM (Interface Homem Máquina) de mais alto nível, num computador pessoal. A interligação do CLP com o computador pessoal é feita via modem através de uma linha telefônica. O sistema foi instalado em uma casa de vegetação do Instituto de Biociências da USP para monitoração e controle da temperatura, da umidade relativa do ar e do fotoperíodo, enquanto o computador de supervisão está localizado no Laboratório de Automação Agrícola, na Escola Politécnica da USP.
Traditional open field agriculture is dependent on the natural environment, and its profit is a result of/ derives from favorable soil, weather and water conditions, among other factors. The increasing need to produce high quality crops, to plan agricultural production in terms of quantity and time, to decrease costs, while maintaining or increasing quality has led to protected agriculture. Agricultural production with some independence of weather conditions can be obtained using greenhouses, which provide good weather protection for the crop. There are many studies showing that the main variables related to crop production are: air temperature, air humidity, solar radiation and carbon dioxide concentration. The maintenance of these variables between a minimum and a maximum limit provides good conditions for crop development, whereas, beyond these limits, the development is restrained. Consequently, a good control policy for these variables is deemed necessary. This work presents Agrilogic, a system for research on greenhouse climate control. It uses industrial automation devices, such as PLC (Programmable Logic Controller), which accounts for short time activities and SCADA (Supervisory Control And Data Acquisition), software responsible for the MMI (Man Machine Interface), which accounts for the long-term activities. The system was installed in a greenhouse at Instituto de Biociências, Universidade de São Paulo for temperature, air relative humidity and photoperiod monitoring and control, and it is linked to a personal computer located at the Agricultural Automation Laboratory, at Escola Politécnica da USP, via a modem and a telephone line.

The purpose of this thesis is to fill the previous research gap concerning automation in the horticulture industry by discovering the adoption of automation in the U.S. and Canada, exploring the possibilities of introducing autonomous solutions and provide recommendations as to how this could create opportunities for small robotics companies targeting the industry. A case company in the U.S. and Canada was used as an example of a small robotics company for the case study. Two research questions were formulated: RQ1: Which major tasks in the horticulture industry should a small robotics company aim to automate? RQ2: What are the barriers for companies in the horticulture industry to invest in automated solutions? A mixed methods research with a pragmatic, inductive and exploratory approach was employed. The primary source of data was gathered from surveys, due to the geographical diversity of the region studied. The surveys reveal that the average level of automation across all respondents averaged at 47%. Given the strategy of the case company, a small robotics company is argued to aim to automate the following tasks: placing plant liners, sticking cuttings and planting seed, spacing of plants and containers, plant pruning, harvesting and grading production, and pesticide application. The horticulture industry is showing low barriers to invest in automation. The relatively high levels of automation are leading to increased trust in automation and further investments in automation. This is shown in the technology being perceived as useful amongst 75-85% of respondents and perceived as easy to use amongst 94% of respondents.
Syftet med denna avhandling är att fylla det tidigare forskargapet om automatisering inom hortikultur, genom att utforska acceptansen av automatisering i USA och Kanada, utforska möjligheterna att införa autonoma lösningar och ge rekommendationer om hur detta kan skapa möjligheter för små robotföretag som riktar sig mot branschen. En fallstudie på ett robotföretag i USA och Kanada användes som ett exempel på ett litet robotföretag. Två forskningsfrågor formulerades: RQ1: Vilka stora uppgifter inom hortikultur bör ett litet robotföretag sträva efter att automatisera?RQ2: Vilka hinder finns för företag inom hortikultur att investera i automatiserade lösningar? En blandad metodforskning med ett pragmatiskt, induktivt och utforskande tillvägagångssätt användes. Den primära källan till data samlades från undersökningar, på grund av den geografiska mångfalden i regionen som studerades. Undersökningarna visar att den genomsnittliga automatiseringsgraden för alla svarande i genomsnitt uppgick till 47%. Med tanke på bolagets strategi rekommenderas ett litet robotföretag att automatisera följande uppgifter: rada upp plantor, stick och plantera frön, skapa avstånd mellan växter och behållare, beskära och kvalitetsgranska skördar, och applicera bekämpningsmedel. Hortikulturindustrin visar låga hinder för investeringar i automatisering. De relativt höga automatiseringsnivåerna leder till ökat förtroende för automatisering och ytterligare investeringar i automation. Detta framgår av tekniken som uppfattas som användbar bland 75–85% av de svarande och uppfattas som lätt att använda bland 94% av de svarande.

The main objectives were: (1) to develop tools to aid in the design of enclosed agro-ecosystems, and (2) to use these tools to develop a prototype simulation-based control system. Three tools were developed: (1) a conceptual framework, (2) a (simulated) greenhouse system and (3) a simulation approach within OS/2.
Part of the conceptual framework was dedicated to "conscious control", defined as a form of control practised by an entity that uses models of itself in its decision-making processes. The greenhouse system was composed of six modules (a simulation manager, a weather generator, a greenhouse model, a crop model, a Pavlovian controller and a cognitive controller), which were implemented under OS/2 as separate processes.
The greenhouse system was used to develop a prototype simulation-based controller. Primarily, the role of the controller was to determine temperature setpoints that would minimize the heating load. The simulation model used by the controller was an artificial neural network. The controller adapted temperature setpoints to anticipated meteorological conditions and reduced greenhouse energy consumption, in comparison with a more traditional controller.
Generally, the results showed the feasibility and illustrated some of the advantages of using simulation-based control. The research resulted in the definition of elements that will allow the creation of a methodological framework for the design of simulation-based control and, eventually, a theory of conscious control.

Os aspectos ambientais vêm ganhando importância para a gestão de grandes, e até mesmo, de pequenas empresas, que passaram a adotar práticas sustentáveis para atender às legislações governamentais, além de buscar eventuais retornos econômicos e mercadológicos. Assim, considerando este cenário, este estudo tem como foco a análise da emissão de gases de efeito estufa, partindo da premissa de que para minimizar esta emissão é necessário examinar todos os processos de extração, produção e transporte ocorridos ao longo da cadeia de suprimentos. Nesse sentido, este estudo tem como objetivo principal o desenvolvimento de uma metodologia para redução da emissão de tais gases em uma cadeia de suprimentos, utilizando tecnologias da automação e de informação. Para verificar sua efetividade, tal metodologia foi aplicada em uma cadeia de suprimentos de etanol, que teve suas emissões estimadas, avaliadas e aperfeiçoadas por meio do uso de um Sistema Inteligente de Navegação, que pretendia diminuir as distâncias rodadas e o consumo de combustível dos veículos usados na distribuição do produto. Os resultados obtidos nas simulações realizadas expõem a possibilidade de reduzir, em um cenário otimista, até 13,42% das emissões da cadeia analisada.
Environmental aspects are gaining importance for the management of big and small organizations. These have adopted environmental practices to meet government laws and to pursue economic and market returns. So, considering this scenario, this study focuses on the analysis of the emission of greenhouse gases, on the premise that to minimize this issue is necessary to examine all processes of extraction, production and transportation occurring along the supply chain. In this sense, this thesis has as main objective the development of a methodology that seeks to reduce the emission of greenhouse gases in a supply chain, using for this technologies of automation and information. To evaluate its effectiveness, the methodology is applied in an ethanol supply chain located in Brazil, which has its emissions estimated, evaluated and possibly optimized through the use of an intelligent navigation system during the distribution process. This technology intends to assist the vehicle\'s driver to find routes that minimize the distance traveled and to maintain an optimal speed set by the manufacturer to reduce the consumption of energy. The results obtained in the simulations show that the implementation of the methodology can reduce, in an optimistic scenario, up to 13.42 % of the greenhouse gases emissions of the ethanol SC analyzed.

Les forages polaires profonds contiennent des enregistrements des conditions climatiques du passé et de l'air piégé qui témoignent des compositions atmosphériques du passé, notamment des gaz à effet de serre. Cette archive nous permet de décrypter le rôle des gaz à effet de serre dans les variations climatiques pendant huit cycles glaciaire-interglaciaires, soit l'équivalent de plus de 800 000 ans. Les carottes de glace, comme toute archive paléoclimatique, sont caractérisées par des incertitudes liées aux processus qui traduisent les variables climatiques en proxy, ainsi que par des incertitudes dues aux chronologies de la glace et des bulles d'air piégées. Nous développons un cadre méthodologique, basé sur la modélisation inverse dite Bayesienne et l'évaluation de fonctions complexes de densité de probabilité, pour traiter les incertitudes liées aux enregistrements paléoclimatiques des carottes de glace de manière précise. Nous proposons deux études dans ce cadre. Pour la première étude, nous identifions les probabilités de localisation des points de changement de pente de l'enregistrement du CO2 dans la carotte de WAIS Divide et d'un stack d'enregistrements de paléotempérature a partir de cinq carottes Antarctiques avec des fonctions linéaires par morceaux. Nous identifions aussi les probabilités pour chaque enregistrement individuel de température. Cela nous permet d'examiner les changements de pente à l'échelle millénaire dans chacune des séries, et de calculer les déphasages entre les changements cohérents. Nous trouvons que le déphasage entre la température en Antarctique et le CO2 à probablement varié (en restant inferieur, generalement, à 500 ans) lors de la déglaciation. L'âge des changements de temperature varie probablement entre les sites de carottage aussi. Ce résultat indique que les mécanismes qui reliaient la température en Antarctique et le CO2 lors de la déglaciation pouvaient être differents temporellement et spatialement. Dans la deuxième étude nous développons une méthode Bayesienne pour la synchronisation des carottes de glace dans le modèle inverse chronologique IceChrono. Nos simulations indiquent que cette méthode est capable de synchroniser des séries de CH4 avec précision, tout en prenant en compte des observations chronologiques externes et de l'information à priori sur les caractéristiques glaciologiques aux sites de forage. La méthode est continue et objective, apportant de la précision à la synchronisation des carottes de glace
Deep polar ice cores contain records of both past climate and trapped air that reflects past atmospheric compositions, notably of greenhouse gases. This record allows us to investigate the role of greenhouse gases in climate variations over eight glacial-interglacial cycles. The ice core record, like all paleoclimate records, contains uncertainties associated both with the relationships between proxies and climate variables, and with the chronologies of the records contained in the ice and trapped air bubbles. In this thesis, we develop a framework, based on Bayesian inverse modeling and the evaluation of complex probability densities, to accurately treat uncertainty in the ice core paleoclimate record. Using this framework, we develop two studies, the first about Antarctic Temperature and CO2 during the last deglaciation, and the second developing a Bayesian synchronization method for ice cores. In the first study, we use inverse modeling to identify the probabilities of piecewise linear fits to CO2 and a stack of Antarctic Temperature records from five ice cores, along with the individual temperature records from each core, over the last deglacial warming, known as Termination 1. Using the nodes, or change points in the piecewise linear fits accepted during the stochastic sampling of the posterior probability density, we discuss the timings of millenial-scale changes in trend in the series, and calculate the phasings between coherent changes. We find that the phasing between Antarctic Temperature and CO2 likely varied, though the response times remain within a range of ~500 years from synchrony, both between events during the deglaciation and accross the individual ice core records. This result indicates both regional-scale complexity and modulations or variations in the mechanisms linking Antarctic temperature and CO2 accross the deglaciation. In the second study, we develop a Bayesian method to synchronize ice cores using corresponding time series in the IceChrono inverse chronological model. Tests show that this method is able to accurately synchronize CH4 series, and is capable of including external chronological observations and prior information about the glaciological characteristics at the coring site. The method is continuous and objective, bringing a new degree of accuracy and precision to the use of synchronization in ice core chronologies

Fotosyntes är central för vår överlevnad. I denna studie har en artificel ljuskälla reglerats för att optimera fotonintensiteten då intensiteten från solen varit för låg. Fokus är lagt på PAR-vågbandet vilket är de våglängder som anses ha störst inverkan på fotosyntes. En kalibrerad sensor användes för att ge pålitliga mätvärden av fotonintensiteten. En billig sensor konstruerades för mätning av densamma och även manuell mätning med lux-meter där värdena konverterades till fotonintensitet. För att studera hur tillväxten påverkas av fotonintensiteten har en testgrupp och en kontrollgrupp med morötter används. Morötterna placerades i ett växthus under liknande förutsättningar med avseende på till exempel näring och värme. Testgruppen fick tillgång till artificellt ljus då intensiteten från solen var för låg under dagstid. Kontrollgruppen fick endast tillgång till solljus. Tillväxten för testgruppen blev större än för kontrollgruppen. För vidare forskning skulle en steglös reglering av fotonbombardemanget kunna optimera processen. Denna förbättring innebär ekonomisk besparing i form av högre biologisk tillväxt vid lägre energiförbrukning. Slutsatsen av denna studie är att optimering av fotonbombardemang kan medföra ekonomisk besparing och även minskad negativ miljöpåverkan på grund av lägre energiförbrukning.
Photosynthesis is central to our survival. In this study, an artificial light source has been regulated to optimize the photon intensity when the intensity from the sun has been too low. The focus is on the PAR waveband, which are the wavelengths that are considered to have the greatest impact on photosynthesis. A calibrated sensor was used to provide reliable measurements of the photon intensity. An inexpensive sensor was designed for measuring the same and also manual measurement with lux-meter where the values ​​were converted to photon intensity. To study how growth is affected by photon intensity, a test group and a control group with carrots were used. The carrots were placed in a greenhouse under similar conditions with regard to, for example, nutrition and heat. The test group gained access to artificial light when the intensity from the sun was too low during the day. The control group only had access to sunlight. The growth for the test group was greater than for the control group. For further research, a stepless regulation of photon bombardment could optimize the process. This improvement means economic saving in the form of higher biological growth at lower energy consumption. The conclusion of this study is that optimization of photon bombardment can lead to economic saving and also reduced negative environmental impact due to lower energy consumption.

碩士
國立成功大學
製造資訊與系統研究所
105
A Greenhouse automatic control and monitoring service system is developed in this paper realized the wireless communication by client-server architecture, we can monitor the environment and control the hardware of the farm remotely by terminator device which via web browser, the data transmission between embedded system and server is realized by the TCP/IP communication protocol. This research mechanism includes remote monitoring cloud service platform, embedded wireless transmission, environment sensing, automatic control system. In environment sensing, we have integrated temperature, humidity, co and soil moisture to building the monitoring system, in automatic control, the automatic irrigation and temperature control system is developed in this paper. And the solar tracking system is developed in this paper to improve the efficiency of green power. This research implemented cloud server software and hardware architecture that saving the cost of the hardware.

碩士
龍華科技大學
電機工程系碩士班
103
Micro Controller Unit was produced in the 1980s. It was researched and developed from General Instrument, its features of low price, multi-function, user-friendly, and its volume is small. Up to now, Micro Controller Unit is be used in many fields, such as automobiles, biomedical engineering, motor controllers, vending machines, sirens, electric vehicle, and so on. Many different types of Micro Controller Unit are promoted continuously. For example: 8051, PIC, HT, and MPC. This essay discusses MSP430 mainly which was developed by Texas Instrument Company. Its major characteristic has a low power design. Moreover, Texas Instrument Company launched many different series of MSP430 and varied packing. MSP430 is highly integrated circuits. It is not only save many peripheral devices but also has many powerful functions. This essay applies the internal ADC12 of MSP430 to transfer analog signals to digital signals of temperature-humidity and illumination so on. Afterwards, we control the MSP430 modules to reveal the messages according to the digital signals from the computer.

M.Ing.
This research project is dedicated to the automation of environmental control within greenhouses. To create an optimal climate in the greenhouse, the main environmental parameters that need to be controlled are temperature, humidity and light intensity. As a result of process dead times and the extreme interdependence of these parameters, the control problem can be classified as non-linear and multi-variable. In the past, most greenhouse environmental control systems depended on the decision making of an experienced human operator. This often gave rise to trial and error, especially when new species were established. With the current advances in "intelligent" control systems and high accuracy sensors, more and more of the decisions involved in greenhouse control can be automated. In this way more emphasis can be placed on emulating the abilities of an expert operator, by means of a computerbased automatic control system. In this research project, "intelligent" as well as "non-intelligent" control techniques, for addressing the problem of automated climate control in a greenhouse, are investigated. These include PID-control as a "non-intelligent" technique, and rule-based fuzzy logic control and self-learning fuzzy logic control as two "intelligent" control techniques. These techniques are all applied to experimental greenhouse which is equipped with management mechanisms, such as fans, heaters, sprinklers and lights. The results of the experiments are evaluated according to two performance parameters: the Control Performance Index (CPI) and the Mean Square Error (MSE). The three techniques are not only assessed for their efficiency, but also for their applicability to the greenhouse environmental problem. Each of the control techniques has a unique characteristic response to the non-linear, non-stationary, multi-variable problem of environmental control and are subsequently addressed in the respective chapter.

碩士
國立中興大學
生物產業機電工程學系所
104
To solve the agricultural damages caused by the climate change, this study is developed to build an automatic supplementary lighting platform for plant factories. The platform mainly uses aluminum as materials. The lighting strategy will depend on the DPI. During the day, platform parked at the original position and uses quantum sensor to accumulate the daylight. At night, it moved to the supplementary lighting area and provides lighting to plants. The platform is operated in cooperation with a PLC control system, and uses RFID and limit switch as a local positioning system at plant factories. In supplementary lighting area, lifting system detects the height of plants to adjust the level of lampshade accordingly. The study operated 10 times movement experiments, after first 5 times calibration, the success rate of last 5 times moving positioning is 93%, and the success rate of 10 times lifting experiments is 98%. The platform can provide an average of 354 μmol /m2 /s light intensity within the distance of 10cm from the plants, and 3.27 μmol CO2 / m2 / s if it is within the photosynthesis characteristic of lettuce. Based on the results of two-week experiment, it can be confirmed that supplementary lighting is able to provide sufficient lighting to plants during the night, and reach DPI standard of each supplementary lighting area.

碩士
國立宜蘭大學
生物機電工程學系碩士班
100
The transporting operation is one of major jobs for culture managements in a greenhouse, however the transporting system is hard to automation because of the space limitation in greenhouses. Therefore, the objective of this study was to develop an automatic guiding vehicle system which could be utilized in transporting operations in a greenhouse. Size of the developed automatic guiding vehicle system was 65 cm in length, 50 cm in width, and 48 cm in height. The moving route of the system utilized virtual tracks to take the place of real gilding objects. The controlling core of the system was a computer; moreover servo motors collocated with motion control cards drove the moving device and precisely controlled movement of the vehicle system which displayed an average moving rate of 0.3 m/s. The system employed a scanning range finder to perform functions of obstacle avoidance and location fixation, while C# programming language was used for software development. In the experimental greenhouse, several immovable benches were built at both sides of the main pavement and had a stationary width of 70 cm between every two benches. Establishing various virtual tracks by track planning in the experimental greenhouse, the meeting positions of various virtual tracks were defined as track points. As both vehicle position and moving target were set, following that the shortest moving track was calculated by a Dijkstra algorithm. Results of track planning test illustrated that the programming track exhibited the shortest route no matter the vehicle moved from the main pavement to side one or from the side pavement to the main one. Capabilities to confirm moving positions was demanded for vehicle movement, therefore, positions locating utilized bench corners as references and track points as revised moving targets. The distance between a track point and the vehicle was calculated and revised by a relative position between a bench corner and a scanning range finder. Generalizing Hough transform was applied to recognize corner shapes of benches, and a 50 × 50 pixels right angle corner was expressed as a scanning image size of 200 × 150 pixels. Results of testing 100 scanning images demonstrated that the successful identification ratio reached 97% and meanwhile the speed of identification was 3 images per second. According to track planning, there were distance errors averaging on 3 cm between tracks points which vehicle actual moved and positions locating by the system. The methods of track planning and position locating used in this research could precisely control the vehicle to move automatically in a greenhouse.

碩士
國立宜蘭大學
生物機電工程學系碩士班
107
The spraying operation in greenhouse is mostly carried out by manually holding spraying device which might cause uneven spraying. Some greenhouses are equipped with top spraying booms. The nozzles on the boom are prone to have the clogging problems. The liquid in the top boom might be heated in summer to cause the damage of the sprayed plants. Therefore, the objective of this study is to develop a spraying system that can automatically spray on Phalaenopsis potted plants in the greenhouse. The system applies machine vision technique to determine the spraying locations of the sphagnum moss in the pots to reduce the waste and pollution of the liquid. A spraying nozzle is installed on the gantry-type moving mechanism driven by a stepping motor. This solenoid nozzle controlled by microprocessor can rotate and quickly spray on pots in the same row based on the image acquired by the camera. The spraying locations are minimized to reduce the stop times of the moving mechanism to enhance the spraying efficiency. An automatic spraying system for Phalaenopsis potted plants has been successfully developed. The experiment result showed that the successful rate of the proposed vision method to determine the spraying location at the different leaf orientations and randomly placed Phalaenopsis is 100%. The successful rate of the vision method for fully filled with potted plants is about 85%. The overall spraying successful rate of the automatic spraying system is about 96.97%.

碩士
建國科技大學
自動化工程系暨機電光系統研究所
103
At present, most of the greenhouses were built with cooling system in Taiwan, making the greenhouse airtight is the only way to decrease the loss of the damages of the cold weather, there is no way to increase the temperature of the greenhouse. The research is for the problems above to develop an automatic cultivation system with green energy integrated on greenhouse. To backup energy with solar energy water heater for greenhouse warming in winter time and to set water wall and/or moisture dispersion sprinkler for greenhouse cooling in summer time. The research and the development of the automatic cultivation system with green energy integrated on greenhouse is divided into three parts, the establishment of the greenhouse, the electric control circuit design, and computer programming for final integration. To experiment the efficiency of heating and cooling systems, to observe the difference of temperature between inside and outside of greenhouse, and to analyze the collected data with for improvement in the future. In order to meet the requirements of the most suitable cultivable environment for plants. This experiment attests that the terms of greenhouse of the automatic cultivation system with green energy integrated and come up with the results, for heating there is 7℃ increased from 29℃ to 36℃, and for cooling there is 5℃ dropped from 33℃ to 27℃～29℃. The system can cultivate plants in four seasons of the whole year with both heating system in the winter and cooling systems in the summer to avoid the damages of the cold weather.

碩士
國立中興大學
農業機械工程學系
87
Abstract The purpose of this study is to develop a computer-based automatic greenhouse environment control system (AGECS) which is simply and easily to operate. To control the greenhouse automatically, the computer controlled programmable logic controller (PLC) and environmental device had been designed and implemented. The model of greenhouse and the graphic control software were used to simulate and compare with an actual one. To construct the values of the luminometer and hygrometer, the analog signals from temperature control module were transduced to digital signals. Consequently, the signals of sensors can be obtained accurately. In this thesis, the Visual Basic (VB) was used to write a communication program so that the PC greenhouse model control and PLC was communicated. Therefore, the human-machine interface was constructed and the purpose of the automatic environmental control for greenhouse can be reached.

碩士
國立暨南國際大學
資訊管理學系
108
In the field of agriculture, evapotranspiration is always a significant indicator for irrigation scheduling. However, to a great extent, it is influenced by the surrounding environment, including illumination, relative humidity (RH), air temperature and wind speed, etc. Since growers typically irrigate crops based on their personal experience, this experience-oriented method cannot cope with the weather in a relatively efficient way and the irrigation water requirement of irrigation is at variance with the crop water requirement. As a result of that, this approach often causes the water waste of irrigation. This study is designed to obtain environmental data during crop growth through the Internet by installing various environmental sensors in the greenhouse. Moreover, based on the environmental data, the evapotranspiration model in the article is able to predict the evapotranspiration of crops which can provide a more accurate crop water requirement for the greenhouse irrigation system, assisting growers to make better judgement according to their irrigation needs. This study is expected to improve the judgment of novice growers on irrigation needs, hoping growers can save more irrigation water and achieve higher efficiency in the greenhouse irrigation system at the same time.

碩士
國立中興大學
資訊科學與工程學系
105
We proposed the automatic pest detection. Main target is to classify and calculate the pest amount of the yellow and blue sticky trap. Therefore, we can gain some information for that. We can control the minimal use of pesticides and higher effect of decreasing pest. Our main target is for the red and brown thrips and white whitefly on the yellow and blue sticky trap. The thrips size is smaller than other pests on the yellow sticky trap and its color is not obvious to others, either. Afterward, let the input images which are converted to YC_b C_r color space from RGB color space. Set the markers when the object has the higher C_b values than the background. Therefore, we use the marker controlled watershed as our segmentation method to separate the pests from the background. Use color and area of segmentation as our features. We use SVM(Support Vector Machine) as our classification. The SVM have the machine learning ability. It is different from some common methods. It is fast and effective to classify the thrips from the segmentations. The part of whitefly, its color is white and have the difference and obvious to the background. Because the size of whitefly is bigger than thrips, the detection and classification is much easier than the thrips. We only use the luminance as the feature to set the markers when the object has the higher luminance values than the background and segment whiteflies on the sticky trap by marker controlled watershed. Therefore, the marker controlled watershed can segment whiteflies well. Finally, we can gain the information of the amount of whitefly to use and analyze.

碩士
國立臺灣大學
生物產業機電工程學研究所
104
Orchids are one of the essential commercial crops in the world. In Taiwan, according to the official statistics, the orchid has a large proportion of the crop planted in export and the orchid export reached 1.83 million USD in 2014. Moreover, orchids are easily damaged by pests, such as dark-winged fungus gnats (Bradysia sp.), resulting in growth retardation or damaged leaves and eventually influencing the price of orchids. To reduce the loss caused by pests and the cost of manual labor, this study developed an automatic monitoring system for pests based on Internet of Things (IoT) to assist floral famers in pest prevention. Two main functions of the proposed system deployed in an orchid greenhouse nursery are environmental monitoring and pest monitoring. In environmental monitoring, a wireless sensor network is responsible for measuring ambient temperature and relative humidity. In pest monitoring, sticky papers with the yellow color of a specially designed wavelength are utilized to attract pests. A camera module is responsible for capturing the images of the papers in a fixed schedule. Furthermore, the images are transmitted to an FTP server. A pest counting algorithm based on light source correction, adaptive binarization, the Canny edge detection and noise reduction is responsible for calculating the number of pests. In addition, the proposed system can automatically change the sticky papers when the number of the pests on the papers exceeds a designed threshold. From Aug. 16 to Nov. 30, 2015, the proposed system has obtained 107 samples, the root mean square error is 2.06, the relative error is 4.91 % ± 0.86 % at a 95 % confidence level. The experimental results show that the counting accuracy of the pest counting algorithm is high. In the future, the sensing data will be utilized to find the relation between the pests and the environment. It is expected that using the proposed monitoring system can provide prescriptive cultivation decisions to farmers based on IoT such that they can prevent pests by using scientific technologies to obtain high yields.

Networked telerobots are operated by humans through remote interactions and have found applications in unstructured environments, such as outer space, underwater, telesurgery, manufacturing etc. In precision agricultural robotics, target monitoring, recognition and detection is a complex task, requiring expertise, hence more efficiently performed by collaborative human-robot systems. A HUB is an online portal, a platform to create and share scientific and advanced computing tools. HUB-CI is a similar tool developed by PRISM center at Purdue University to enable cyber-augmented collaborative interactions over cyber-supported complex systems. Unlike previous HUBs, HUB-CI enables both physical and virtual collaboration between several groups of human users along with relevant cyber-physical agents. This research, sponsored in part by the Binational Agricultural Research and Development Fund (BARD), implements the HUB-CI model to improve the Collaborative Intelligence (CI) of an agricultural telerobotic system for early detection of anomalies in pepper plants grown in greenhouses. Specific CI tools developed for this purpose include: (1) Spectral image segmentation for detecting and mapping to anomalies in growing pepper plants; (2) Workflow/task administration protocols for managing/coordinating interactions between software, hardware, and human agents, engaged in the monitoring and detection, which would reliably lead to precise, responsive mitigation. These CI tools aim to minimize interactions’ conflicts and errors that may impede detection effectiveness, thus reducing crops quality. Simulated experiments performed show that planned and optimized collaborative interactions with HUB-CI (as opposed to ad-hoc interactions) yield significantly fewer errors and better detection by improving the system efficiency by between 210% to 255%. The anomaly detection method was tested on the spectral image data available in terms of number of anomalous pixels for healthy plants, and plants with stresses providing statistically significant results between the different classifications of plant health using ANOVA tests (P-value = 0). Hence, it improves system productivity by leveraging collaboration and learning based tools for precise monitoring for healthy growth of pepper plants in greenhouses.