Objective of this dissertation was to study possibilities of detection of subsurface drainage systems by unmanned aerial vehicles. Unmanned aerial vehicles, "drones", served to military purpose in the past. Nowadays these unmanned aerial vehicles helps to map drainage on fields because it is hard to get draining plans today and a big part do not match with the actual implementation of drainage systems. We can also add that nobody cares about drainage today. Drainage drains of soaked soil and improve its harvest. Drainage facilitates transport of nutrients (nitrogen and phosphorus) to watercourses. Fields changes over time to parcels. There is description of sorting and functionality of drainage systems as well as description of malfunctions and used material in theoretical part. There is also description of unmanned aerial vehicles and a method of detection of subsurface drainage systems . Besides there is detailed description of the locality, function of drainage system and evaluation. There is description of indication of drainage on specialized aerial photography. Indications are compared with period plans. Methods of highlighting indications are applied and evaluated by programs Agisoft PhotoScan Professional and Arc GIS 10.1. Main results are photographies that demonstrate location of drainages. Using ground control points the image converted into the coordinate system.

La collecte de données par des réseaux de capteurs autonomes mobiles peut être couplée à l’utilisation de drones qui constituent une solution de backahauling facilement déployable à faible coût. Ces moyens de collecte peuvent servir lors de l’organisation d’évènements temporaires (sportifs ou culturels) ou encore pour mener des opérations dans des terrains difficiles d’accès ou hostiles. L’objectif de cette thèse est de proposer des solutions efficaces pour la communication à la fois entre capteurs mobiles au sol et sur la liaison bord-sol. A ces fins, nous nous intéressons à l’ordonnancement des communications, au routage et au contrôle de l’accès sur la liaison capteurs/drone, le collecteur mobile. Nous proposons une architecture répondant aux contraintes du réseau. Les principales sont l’intermittence des liens et donc le manque de connexité pour lesquelles des solutions adaptées aux réseaux tolérants aux délais sont adoptées. Vu la limitation des opportunités de communication avec le drone et la variation importante du débit physique, nous avons proposés des solutions d’ordonnancement qui tiennent compte à la fois des durées de contact que du débit physique. Le routage opportuniste est également fondé sur ces deux critères à la fois pour la sélection des nœuds relais que pour la gestion des files d’attente. Nous avons souhaité limiter l’overhead et proposer des solutions efficaces et équitables entre capteurs mobiles au sol. Les solutions proposées ont montré leur supériorité par rapport aux solutions d’ordonnancement et de routage classiques. Nous avons enfin, proposé une méthode d’accès combinant un accès aléatoire avec contention ainsi qu’un accès avec réservation tenant compte des critères précédemment cités. Cette solution flexible permet à un réseau de capteurs mobiles denses de se rapprocher des performances obtenues dans un mode oracle. Les solutions proposées peuvent être mises en œuvre et appliquées dans différents contextes applicatifs pour lesquels les nœuds au sol sont mobiles ou aisément adaptées au cas où les nœuds sont statiques. / Data collection by autonomous mobile sensor arrays can be coupled with the use of drones which provide a low-cost, easily deployable backhauling solution. These means of collection can be used to organize temporary events (sporting or cultural) or to carry out operations in difficult or hostile terrain. The aim of this thesis is to propose effective solutions for communication between both mobile sensors on the ground and on the edge-to-ground link. For this purpose, we are interested in scheduling communications, routing and access control on the sensor / drone link, the mobile collector. We propose an architecture that meets the constraints of the network. The main ones are the intermittence of the links and therefore the lack of connectivity for which solutions adapted to the networks tolerant to the deadlines are adopted. Given the limited opportunities for communication with the drone and the significant variation in the physical data rate, we proposed scheduling solutions that take account of both the contact time and the physical flow rate. Opportunistic routing is also based on these two criteria both for the selection of relay nodes and for the management of queues. We wanted to limit the overhead and propose efficient and fair solutions between mobile sensors on the ground. The proposed solutions have proved superior to conventional scheduling and routing solutions. Finally, we proposed a method of access combining a random access with contention as well as an access with reservation taking into account the aforementioned criteria. This flexible solution allows a network of dense mobile sensors to get closer to the performance obtained in an oracle mode. The proposed solutions can be implemented and applied in different application contexts for which the ground nodes are mobile or easily adapted to the case where the nodes are static.

Dissertation (MTech (Mechanical Engineering))--Cape Peninsula University of Technology, 2011 / This paper presents the development of a data collection system for a small unmanned Aerial Vehicle (UAV) flight. The following three facets comprise of a UAV system: (1) a UAV aircraft; (2) onboard avionics; and (3) a ground control station subsystem (Taha et al., 2010:1). In this project, the UAV aircraft is based on the low-cost autonomous quad-rotator system named “Arducopter Quad”, where the onboard avionic system utilizes both an ArduPilot Mega (APM) on-board controller and IMU sensor shield, while the “Mission Planner” software operates as GCS software to gather essential flight data (Xiang & Tian, 2011:176). The approach provides the UAV system structure and both hardware and software with a small UAV data collection system, which is examined throughout the study. And introduce the concept of Arducopter dynamics for better understanding with its flight control. The study also considers the communication process between the UAV and the ground control station. The radio wave is an important aspect in the UAV data collection system (Austin, 2010:143). The literature review introduced the basis of the radio wave in respect of its travelling speed, and its characteristics of propagation, including how different frequencies will affect radio wave propagation. The aim of this project was to develop a platform for a small UAV real-time data collection system. The pendulum system was involved to simulate the “Roll” movement of the small UAV, while real-time IMU sensor data was successfully collected at ground control station (GCS), both serial communication and wireless communication, which was applied in the data collection process. The microwave generator interference test proves that the 2.4 GHz XBee module is capable of establishing reliable indoor communication between the APM controller and the GCS. The work of this project is towards development of additional health monitoring technology to prevent the safety issue of the small UAV. The data collection system can be used as basis for the future research of real-time health monitoring for various small UAVs.

In order to successfully compete against the use of fossil fuels to generate electricity, one of the challenges in the photovoltaic (PV) business currently in focus is on the asset management of large PV plants, in which developing control techniques to prognosticate and evaluate the future energy performance will be essential. Infrared thermography inspections can give meaningful support to assess the quality and performance of PV modules. However, the implementation of a cost-effective method to scan and check huge PV plants represents different challenges, such as the cost and time of detecting PV module defects with their classification and exact localization within the solar plant. In this context, it has recently been investigated the potential of a new innovative technology in the PV plants monitoring operations by using drones. The main purpose of this work is to establish a scientific basis for the interpretation of thermographic images taken by drones, in particular, regarding the influence of thermographic irregularities which will negatively influence the performance of PV plants. The drone is employed to monitor PV modules conditions by using special thermography sensors mounted on it in order to scan images. The captured images are then automatically sent to a technical office database for the image processing software. This special software receives, stores and analyses the captured images to detect the specific defect on the PV modules. Then, all information is processed and reported to the final decision-making team to decide about the best solution for the particular degraded PV module, in relation with the requirements from the operation and maintenance (O&M) services. In this particularly study project of the inspected PV plant situated in the UK, which has been carried out by trained personnel at Quintas Energy (QE), the majority of identified faults, which influence the PV module performance (especially the power output significantly), are on a sub-panel level, either individual cells or uneven hot spots. There are also some modules with bypass diode faults as well as a string fault was detected. Such faults must be repaired by the PV module manufacturer, in relation to the manufacturer’s warranties, without any cost at all since the PV modules are indeed still in warranty. It has been concluded that, in comparison with traditional manned systems by using hand-held cameras, the main functionality of using drones is the early fault diagnosis which could reduce corrective maintenance activities, since defects are easily and quickly identified and, then, repaired. This fact could reduce defects to become more serious and, thus, more difficult to be repaired, along with their correspondent production losses and costs. QE has learned by making mistakes during this project study and gained experience of this unmanned aerial vehicles (UAV) technology. Currently, they are in the process of improving this technique and will continue to implement it to all their PV plants since the efficiency of PV systems can be significantly improved by appropriate use of O&M instruments and benefit from innovative monitoring tools, such as the unmanned aerial technology.

Este trabalho dedica-se ao projeto e desenvolvimento de um veículo aéreo não tripulado. Tais veículos podem ser utilizados em diversas aplicações como monitoramento, vigilância, transporte, resgate, entre outros. Dentre os diversos tipos de veículos aéreos, este trabalho irá focar no modelo do quadrirotor, composto por quatro hélices contra-rotoras que estabilizam e movimentam o veículo. Para alcançar o objetivo de controlar este tipo de veículo, várias propostas e metodologias podem ser aplicadas, todas buscando contemplar o controle de todas ou parte das variáveis de estado presentes nesta planta. Neste texto serão descritas: as equações cinemáticas e dinâmicas que regem este sistema; o projeto e composição mecânica da aeronave; definição de sensores e atuadores juntamente com seus métodos de utilização; implementação de controlador linear por alocação direta de polos e Regulador Linear Quadrático juntamente com observador de estados de ordem plena e filtro de Kalman, para recuperação de estados não mensurados e filtragem de ruídos. Serão apresentados resultados em simulações para cada método de controle selecionado visando optar pelo melhor controlador para a aplicação da aeronave. O método selecionado será implementado para controlar a aeronave com os sensores e atuadores selecionados. Esta implementação será realizada a partir da técnica HIL Hardware in The Loop juntamente com o software MATLAB/Simulink visando validar o controlador em conjunto com a planta real, bem como o modelo dinâmico construído. / This work is dedicated to the project and development of an unmanned aerial vehicle. Such vehicles can be employed in various applications such as monitoring, surveillance, transportation, rescue and others. Among the types of aerial vehicles, this work is focused on the quadrotor, composed by four counter-rotating propellers which stabilize and displace the vehicle. In order to fulfill the objective of controlling this vehicle, many methodologies and propositions can be applied, seeking the control of all or a snippet of the state variables present in the system. There will be described in this work: the cinematic and dynamic equations that govern this system; the mechanical project and construction of the aircraft; sensors and actuators definition, along with its usage methods; linear control implementation of the pole placement and Linear Quadratic Regulator techniques along full order state observer and Kalman filtering in order to recover and filter non-measured states. Performance results in simulations will be presented on each control implementation to validate the best controller for the application and this implementation will be applied on the projected aircraft using the sensors and actuators selected. This implementation will be given through the HIL - Hardware in the Loop method using MATLAB/Simulink software to validate the control technique applied and the constructed dynamic model.

Sistemas de Transportes Inteligentes (STIs) englobam um conjuntos de tecnologias (Sensoriamento Remoto, Tecnologia da Informação, Eletrônica, Sistemas de Comunicação de Dados entre outros) que visam oferecer serviços e gerenciamento de tráfego avançado para meios de transporte rodoviário, aéreo e outros. A obtenção de informações a respeito das características e das condições do pavimento das estradas constitui uma parte importante dentro do sensoriamento nesses STIs. Investigar novas técnicas, metodologias e meios de automatizar a obtenção dessas informações é parte deste trabalho. Uma vez que existem diferentes tipos de defeitos em vias pavimentadas, esta tese apresenta a proposta de uma metodologia que permite a obtenção, de forma automática, das condições dos pavimentos asfálticos. A obtenção dos dados foi realizada por meio do Sensoriamento Remoto com uso de Veículos Aéreos Não Tripulados. A utilização de técnicas de Aprendizado de Máquina na detecção automática possibilitou alcançar uma acurácia de 99% na detecção de pavimentos asfálticos flexíveis e 92% na identificação de defeitos em alguns experimentos. Como resultado obteve-se o diagnóstico automático, não só das condições da via, mas de diferentes tipos de defeitos presentes em pavimentos. / Intelligent Transport Systems (ITS) is a set of integrated technologies (Remote Sensing, Information Technology, Electronics, Data Communication Systems among others) that aims to provide services and advanced traffic management for road, air, rail and others transportation systems. Obtaining information about characteristics and road pavement conditions is an important part within the sensing these ITS. Investigating new techniques, methods and means to optimize and automate obtaining these information are part of this work, since there are different types of defects on paved roads. Thus, this thesis proposes a methodology that allows automatically obtain information about the condition of the pavement. Data collection was performed with remote sensing technology using Unmanned Aerial Vehicles. Automatic detection was possible through the use of Machine Learning techniques with 99% of accuracy in pavements and 92% in distress identification. As a result we obtained the self-diagnosis, not just the pavement, but different types of distress present in the pavement.

L'objet de cette thèse est de démonter la faisabilité de conception d'un mini-drone longue endurance sans recourir à des véhicules de grande envergure qui nécessite des infrastructures supplémentaires, des systèmes de lancement complexes et un personnel d'exploitation important. Pour ce faire, une approche d'optimisation globale du problème a été utilisée, en s'appuyant sur les spécificités de chacun des aspects de la conception de mini-drones. Ce concept de mini-drone longue endurance doit repousser les limites dans plusieurs disciplines telles que l'aérodynamique, la propulsion, les structures, les sources d'énergies et le stockage, le contrôle et la navigation, ainsi que la miniaturisation de l'électronique embarquée. Un programme de conception baptisé Cdsgn a été développé et prend en compte les problèmes spécifiques de chaque discipline consacrées aux mini-drones. Il permet de voir l’influence de chaque paramètre de conception sur la performance finale de la conception, menant à la sélection optimale des paramètres. Cdsgn génère et analyse rapidement de nombreuses configurations de l'avion tout en simulant la performance de chaque configuration pour un pro fil de mission donnée. Un outil de post-traitement a également été développé afin de filtrer et sélectionner de manière interactive les paramètres de conception parmi les nombreuses configurations pour répondre à des applications pratiques. Le programme proposé a été utilisé dans le développement et la conception de plusieurs projets, tels que Solar Storm, premier mini-drone hybride au monde à énergie solaire d'une envergure de cinquante centimètre, SPOC, un mini-drone longue distance conçu pour voler au-dessus de la mer Méditerranée de Nice jusqu'en Corse (Calvii) et enfin Eternity, mini-drone de longue endurance d'une envergure d'un mètre, avec une configuration classique. Capable d'une autonomie de quatre heures avec les batteries embarquées, son temps de vol peut être amélioré jusqu’à huit heures avec l'utilisation de l’énergie solaire. En utilisant les évaluations de chaque projet, Cdsgn a été amélioré à la fois pour l'exactitude des calculs et pour la performance opérationnelle afin de développer le plus petit véhicule aérien pour une mission d'endurance donnée. / This thesis shows the feasibility of designing a long endurance mini UAV without resorting to large scale vehicles which requires additional infrastructure, complex launching systems and numerous operating crew. To do so is possible by using a global optimisation approach concentrated specifically on each aspect of the mini-UAV design with their particular challenges. So called Long Endurance Mini UAV Concept has to push the limits in several disciplines such as aerodynamics, propulsion, structures, energy source and storage, control and navigation, miniaturised electronics. A conceptual design program called Cdsgn is developed which takes into account each discipline's specific problems devoted to mini UAVs and making it possible to see the influence of each design parameter on the final performance of the complete design leading to the optimum selection of parameters. Cdsgn generates and analyse numerous aircraft configurations rapidly while simulating the performance of each configuration for a given mission profile. A post processing tool is also developed in order to interactively filter and select the final design parameters among numerous analysed aircraft configurations for practical applications. The proposed program is used in the development and design of several projects, such as Solar Storm, the world's first hybrid solar powered micro UAV in half a meter scale, SPOC, a long range mini UAV which is designed to fly across the Mediterranean sea from Nice to Corsica (Calvi) and finally the Eternity, the long endurance mini UAV concept which is an electrically powered, one-meter span aircraft with a conventional configuration having an endurance of four hours with the on-board batteries which can be enhanced up to eight hours with the use of solar-cells. Using the feedback of each project, Cdsgn has been improved both for the accuracy and for the operational performance in order to develop the smallest aerial vehicle for a given endurance mission.

A aplicação de defensivos químicos em áreas agrícolas é de primordial importância para o rendimento de lavouras. O uso de aeronaves é cada vez mais comum em tal tarefa, principalmente pelo fato de aumentar a agilidade na operação. Entretanto, características climáticas, como intensidade e orientação do vento, podem causar prejuízo aos produtores (e.g. aplicação sobreposta dos defensivos e multas por aplicação de defensivos na borda exterior da lavoura). Essa lacuna motivou este trabalho de mestrado, que tem por objetivo avaliar um sistema integrado de veículos aéreos não tripulados (VANTs) e redes de sensores sem fio (RSSFs) para aumentar a eficiência da aplicação de defensivos químicos em campos agrícolas. Um VANT é utilizado para percorrer o campo enquanto aplica defensivos químicos. Uma RSSF com nós capazes de sensoriar a concentração de defensivos químicos é instalada no solo do campo. Os nós da RSSF e o VANT são equipados com módulos de rádio, possibilitando a troca de informação entre eles. Desse modo, o VANT consegue obter a distribuição de defensivos químicos aplicados no campo, podendo então tomar decisões para aumentar a eficiência da aplicação. Para análise desse sistema, dois experimentos foram realizados utilizando o simulador OMNeT++ e o framework MiXiM. Em um primeiro experimento é comparada a eficiência do sistema proposto com o sistema tradicional de pulverização agrícola, mostrando que é possível aumentar, em média, 16% a quantidade de defensivos químicos aplicados dentro do campo agrícola. E, em alguns casos, é possível obter um aumento de até 118,25%. No segundo experimento são analisados cinco diferentes protocolos de comunicação, mostrando que é possível otimizar o uso da bateria nos nós sensores sem fio em até 10 vezes, sem diminuir a eficiência do sistema / The use of chemical defensives in crop fields is important for the agricultural production. Often airplanes are used for spraying these products, because it adds agility in such task. But climatic characteristics, such as wind speed, may cause loss to farmers (e.g. superimposed application and fines for environment contamination). This text details an integrated system of unmanned aerial vehicle (UAV) and wireless sensor network (WSN) with the purpose of letting the aerial spraying process more precise. An UAV is used to spray chemicals in the crop field. A WSN composed of sensor nodes capable of sensoring quantities of chemical defensives is installed in the crop. Both UAV and sensor nodes are equipped with radio modules, this way they can communicate each other. Thus, the UAV may obtain the information of where the chemicals are being applied, then the UAV may take some action to performs a better spraying. This system were developed in the simulator OMNET++ using the framework MiXiM and two experiments were executed. In the first experiment the objective is to compare the efficiency of this system with the tradicional system, without the WSN. Results shown that the proposed system can improve the spraying process in 16% on average, and 118.25% on some cases. In the second experiment, five different WSN protocols were evaluated in means of battery usage of the sensor nodes. Results shown that its possible to reduce the battery usage in 10 times without intervene in the system efficiency

Veículos Aéreos não Tripulados (VANTs) são aeronaves que voam sem tripulação e são capazes de realizar diversos tipos de missões, como vigilância, coleta de dados topográficos e monitoramento ambiental. Este é um domínio que tem muito a ganhar com a aplicação da abordagem de Linha de Produtos de Software (LPS), uma vez que é rico em variabilidades e cada modelo de VANT tem também muitas partes comuns. Neste trabalho é apresentada uma infraestrutura tecnológica e de configuração de ativos em Simulink, gerenciados pelas ferramentas Pure::variant e Hephaestos para uma LPS de VANTs. Um conjunto de padrões para especificação de variabilidades em Simulink é proposto, bem como uma extensão para a ferramenta Hephaestus. Uma comparação entre as ferramentas Pure::variants e Hephaestus é apresentada / Unmanned Aerial Vehicles (UAVs) are aircrafts that can fly without any crew and are capable to realize several types of missions such as surveillance, topographic data collection and environmental monitoring. This is a domain which can benefit very much with the adoption of the Software Product Lines (SPL) approach, as each UAV model is rich in variabilities and has many common parts. In this work it is presented a software asset configuration infrastructure for the Simulink environment, managed by the tools Pure::variants and Hephaestus for a UAV SPL. A set of patterns of variability specification in Simulink is proposed as well as an extension to Hephaestus to support a SPL product engineering for Simulink. A comparison between Pure::variants and Hephaestus is also presented

No description available.