Thesis (MScEng)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: This project concerns motion planning for a rotary wing UAV, where vehicle controllers are already in place, and map data is readily available to a collision detection module. In broad terms, the goal of the motion planning algorithm is to provide a safe (i.e. obstacle free) flight path between an initial- and goal waypoint. This project looks at two specific motion planning algorithms, the Rapidly Exploring Random Tree (or RRT*), and the Probabilistic Roadmap Method (or PRM). The primary focus of this project is learning how these algorithms behave in specific environments and an in depth analysis is done on their differences. A secondary focus is the execution of planned paths via a Simulink simulation and lastly, this project also looks at the effect of path replanning. The work done in this project enables a rotary wing UAV to autonomously navigate an uncertain, dynamic and cluttered environment. The work also provides insight into the choice of an algorithm for a given environment: knowing which algorithm performs better can save valuable processing time and will make the entire system more responsive. / AFRIKAANSE OPSOMMING: ’n Tipiese vliegstuuroutomaat is daartoe in staat om ’n onbemande lugvaartvoertuig (UAV) so te stuur dat ’n stel gedefinieerde punte gevolg word. Die punte moet egter vooraf beplan word, en indien enige verandering nodig is (bv. as gevolg van veranderinge in die omgewing) is dit nodig dat ’n menslike operateur betrokke moet raak. Vir voertuie om ten volle outonoom te kan navigeer, moet die voertuig in staat wees om te kan reageer op veranderende situasies. Vir hierdie doel word kinodinamiese beplanningsalgoritmes en konflikdeteksiemetodes gebruik. Hierdie projek behels kinodinamiese beplanningsalgoritmes vir ’n onbemande helikopter, waar die beheerders vir die voertuig reeds in plek is, en omgewingsdata beskikbaar is vir ’n konflikdeteksie-module. In breë terme is die doel van die kinodinamiese beplanningsalgoritme om ’n veilige (d.w.s ’n konflikvrye) vlugpad tussen ’n begin- en eindpunt te vind. Hierdie projek kyk na twee spesifieke kinodinamiese beplanningsalgoritmes, die “Rapidly exploring Random Tree*” (of RRT*), en die “Probabilistic Roadmap Method” (of PRM). Die primêre fokus van hierdie projek is om die gedrag van hierdie algoritmes in spesifieke omgewings te analiseer en ’n volledige analise te doen op hul verskille. ’n Sekondêre fokus is die uitvoering van ’n beplande vlugpad d.m.v ’n Simulink-simulasie, en laastens kyk hierdie projek ook na die effek van padherbeplanning. Die werk wat gedoen is in hierdie projek stel ’n onbemande helikopter in staat om outonoom te navigeer in ’n onsekere, dinamiese en besige omgewing. Die werk bied ook insig in die keuse van ’n algoritme vir ’n gegewe omgewing: om te weet watter algoritme beter uitvoertye het kan waardevolle verwerkingstyd bespaar, en verseker dat die hele stelsel vinniger kan reageer.

abstract: The objective of this work is to develop a Stop-Rotor Multimode UAV. This UAV is capable of vertical take-off and landing like a helicopter and can convert from a helicopter mode to an airplane mode in mid-flight. Thus, this UAV can hover as a helicopter and achieve high mission range of an airplane. The stop-rotor concept implies that in mid-flight the lift generating helicopter rotor stops and rotates the blades into airplane wings. The thrust in airplane mode is then provided by a pusher propeller. The aircraft configuration presents unique challenges in flight dynamics, modeling and control. In this thesis a mathematical model along with the design and simulations of a hover control will be presented. In addition, the discussion of the performance in fixed-wing flight, and the autopilot architecture of the UAV will be presented. Also presented, are some experimental "conversion" results where the Stop-Rotor aircraft was dropped from a hot air balloon and performed a successful conversion from helicopter to airplane mode. / Dissertation/Thesis / M.S.Tech Mechanical Engineering 2011

A mitigação de desastres naturais exige respostas rápidas e confiáveis. No Brasil, a estação de chuvas provoca muitos alagamentos em regiões urbanas e, para monitorar esse fenômeno, foi instalada em São Carlos-SP uma rede de sensores sem fio para acompanhar o nível de água dos rios da cidade. Entretanto, essa rede de sensores está suscetível a falhas que podem comprometer o funcionamento do sistema, e a adoção de mecanismos redundantes e de redes móveis 3G podem acarretar em custos proibitivos à monitoração desses rios, além de não garantirem a operação normal desse monitoramento. Assim, este trabalho apresenta uma solução baseada em veículo aéreo não tripulado (VANT) para reduzir os problemas oriundos das falhas em uma rede de sensores para detectar desastres naturais como enchentes e deslizamentos. Na solução proposta, o VANT pode ser transportado para o sítio do desastre para minimizar os problemas provenientes das falhas (por exemplo, para servir como um roteador ou até mesmo para servir como uma mula de dados e transmitir imagens em tempo real para equipes de resgate). Estudos foram conduzidos em um protótipo real, equipado com o UAV Brain (módulo computacional desenvolvido especificamente para este projeto), para uma análise exploratória do consumo energético do VANT e do rádio transmissor que equipa o VANT. Os resultados mostram que a melhor situação para o rádio comunicador se dá quando o VANT está no ar e com uma antena de maior ganho, e os fatores que mais influenciam no consumo energético do rádio são a altura do VANT e o tipo de antena utilizado. Além disso, tais resultados mostram também a viabilidade desta proposta em redes de sensores sem fio linear e esparsa. / The mitigation of natural disasters requires quick and reliable answers. In Brazil, the rainy season causes many flooding in urban areas and, to monitor this phenomenon, a wireless sensor network to monitor the water level of the citys rivers was installed in São Carlos-SP. However, this sensor network is susceptible to failures that may jeopardize the operation of the system, and the adoption of redundant mechanisms and 3G mobile networks may result in prohibitive costs to the monitoring of these rivers, and does not guarantee the normal operation of monitoring. This work presents an aerial vehicle-based solution unmanned (UAV) to reduce the problems originated from failures in a network of sensors to detect natural disasters such as floods and landslides. In the proposed solution, the UAV can be transported to the disaster site to minimize problems arising from failures (eg, to serve as a router or even to serve as a data mule and transmit images in real time for rescue teams). Studies were conducted in a real prototype, equipped with UAV Brain (computational module developed specifically for this project), for an exploratory analysis of the energy consumption UAV and radio transmitter fitted to the UAV. The results show that the best situation for the radio communicator is when the UAV is in the air and with a higher gain antenna, and the factors that most influence on the radio energy consumption are the height of the UAV and the type of antenna used. Furthermore, these results also show the feasibility of this proposal in linear and sparse wireless sensor networks.

Au cours de la dernière décennie, les drones ont été largement utilisés dans les domaines des applications civiles. La photogrammétrie aéroportée a trouvé place dans ces applications comme une solution efficace de modélisation 3D mais aussi comme un outil de mesure. Vinci-Construction-Terrassement est une entreprise privée spécialisée dans le secteur des Travaux Publics qui intègre les drones et la photogrammétrie comme une solution de cartographie et de métrologie de ses chantiers. Cet outil est très efficace, par exemple, pour le calcul des volumes de stocks ou pour le suivi temporel de zones spécifiques avec un risque de glissement de terrain. Le but de ce travail est d’arriver à un géo-référencement direct des images acquises par la caméra lors du vol en s’appuyant uniquement sur un récepteur GPS embarqué. Le système utilisé doit être de faible coût et par conséquent le traitement des données est adapté à cette contrainte / Over the last decade, drones have been largely used for civil applications. Airborne photogrammetry has found place in these applications as a modeling and a measuring tool. Vinci-Construction-Terrassement is a private company of public building and works sector that integrates drones and photogrammetry as a mapping solution and metrology investigation on its sites. This tool is very efficient for the calculation of stock volumes for instance, or for time tracking of specific areas with risk of landslides. The aim of the present work is to do direct georeferencing of images acquired by the camera leaning on an embedded GPS receiver. The UAS used needs to be low cost and therefore data processing is adapted to this constraint

Unmanned Aerial Systems (UAS) are becoming more widely used in the new era of evolving technology; increasing performance while decreasing size, weight, and cost. A UAS equipped with a Flight Control System (FCS) that can be used to fly semi- or fully-autonomous is a prime example of a Cyber Physical and Safety Critical system. Current Cyber-Physical defenses against malicious attacks are structured around security standards for best practices involving the development of protocols and the digital software implementation. Thus far, few attempts have been made to embed security into the architecture of the system considering security as a holistic problem. Therefore, a Hierarchical, Embedded, Cyber Attack Detection (HECAD) framework is developed to provide security in a holistic manor, providing resiliency against cyber-attacks as well as introducing strategies for mitigating and dealing with component failures. Traversing the hardware/software barrier, HECAD provides detection of malicious faults at the hardware and software level; verified through the development of an FPGA implementation and tested using a UAS FCS.

This master thesis report presents the developement of an INS/GPS navigation loop written in ANSI C++ using a standard matrix library. The filter have been tested on an Unmanned Aerial Vehicle (UAV) called Brumby. Here data have been logged from the Inertial Measurement Unit (IMU) and the Global Positioning System (GPS) receiver. This data have then been postprocessed and run through the navigation filter for estimation of position, attitude and velocity of the vehicle during the flights. The error feedback to the Inertial Navigation System (INS) is done with a complement filter implemented using a kalman filter written in information form.The resulting navigation filter estimates the attitude within two degrees with 95% confidence and the position within two meter using 95% confidence. / <p>Validerat; 20101217 (root)</p>

There is a large number of design options to consider when designing aircraft vehicle systems for fighter aircraft, and there is a lack of tool support that provides an overview of these available design options. Various design options will bring consequences in terms of weight, performance, cost, etc. which is desired to be known in an early conceptual phase. Conventional methods, such as morphological matrix and design structure matrix, lack the ability to generate an overview and map complex systems. By studying model-based tools in form of ontologies and feature models in Protégé and FeatureIDE respectively, these tools are considered to provide a higher level of detail regarding the available design options compared to the conventional methods, such as the morphological matrix and the design structure matrix. Ontologies and feature-models are therefore considered to increase the effectiveness in the conceptual design phase of aircraft vehicle systems. By combining ontologies and feature models, more thoughtful design decisions can be performed. An increased knowledge of the available design options can lead to an improved development of aircraft vehicle systems, and new solutions can be evaluated. By performing more detailed trade studies for an unmanned aerial vehicle, for different system solutions, various parameters such as engine power outtake, system weight, etc. can be analysed and provide an indication whether a concept should be evaluated further. It is however necessary to consider how different parameters affect the overall system, and fuel penalty may be implemented as an equivalent parameter. Performing power flow calculations do however not consider solution-specific limitations, which have to be implemented to be able to determine if an aircraft vehicle system concept should be considered advantageous or not.

Feature detection and matching is an important step in many object tracking and detection algorithms. This paper discusses methods to improve upon previous work on the SYnthetic BAsis feature descriptor (SYBA) algorithm, which describes and compares image features in an efficient and discreet manner. SYBA utilizes synthetic basis images overlaid on a feature region of interest (FRI) to generate binary numbers that uniquely describe the feature contained within the FRI. These binary numbers are then used to compare against feature values in subsequent images for matching. However, in a non-ideal environment the accuracy of the feature matching suffers due to variations in image scale, and rotation. This paper introduces a new version of SYBA which processes FRI’s such that the descriptions developed by SYBA are rotation and scale invariant. To demonstrate the improvements of this robust implementation of SYBA called rSYBA, included in this paper are applications that have to cope with high amounts of image variation. The first detects objects along an oil pipeline by transforming and comparing frame-by-frame two surveillance videos recorded at two different times. The second shows camera pose plotting for a ground based vehicle using monocular visual odometry. The third generates panoramic images through image stitching and image transforms. All applications contain large amounts of image variation between image frames and therefore require a significant amount of correct feature matches to generate acceptable results.

Unmanned Aerial Vehicle (UAV) är en benämning på en obemannad flygande farkost. UAV är en benämning för själva farkosten och därför har Unmanned Aircraft System (UAS) tagit över eftersom det är ett begrepp som rör hela systemet som förutom flygfarkost innefattar start, landning, markstation och kommunikationslänk. Inom mätningsteknik är UAS ett relativt nytt begrepp och tekniken har sin historia mestadels inom det militära området. Syftet med denna studie är att analysera samt utvärdera två programvaror och en molntjänst för bearbetning och framtagning av ortofoto från UAS-data. De frågor som ställts inför arbetet är: kan en molntjänst ersätta ett avancerat datorprogram vid generering av ortofoton? Kan dessa datorprogram ge ett bra resultat utan hjälp av andra GIS-program? Vilket program är enklast att förstå och använda samt vilka är skillnaderna mellan programmen? Dessa frågor har besvarats genom användning av insamlat data och för att få utvärderingen rättvis har därför tre olika dataset skapats. Programtjänsterna som har utvärderats är Agisoft PhotoScan 0.9.0 och Pix4UAV Desktop/Cloud 2.1.2. Insamling av data har skett genom en flygning med en oktokopter över Fågelmyratippen i Dalarna. Resultaten visar att priset snabbt blir högt om endast Pix4UAV Cloud används och att överlag är PhotoScan billigare än Pix4UAV Desktop. Kvalitetsrapporten som följer med varje projekt är överskådlig i PhotoScan och mer ingående i Pix4UAV Desktop/Cloud. Trots samma indata blir utdatat olika vid bearbetning av de olika programmen, till exempel skiljer sig markupplösningen åt mellan programmen. Generellt är PhotoScan tydligare på att visa hur arbetsprocessen går till. Supporten hos båda företagen är bra, tips och tricks finns på respektive hemsida. Till PhotoScan finns även en manual för nedladdning samt en YouTube-kanal med instruktionsvideor. De enda slutsatserna vi drar är att Pix4UAV Cloud inte klarar av att ersätta ett avancerat bildbehandlingsprogram och att för tillfället bör ytterligare ett GIS-program användas som komplement för att få bästa resultat. I övrigt har vi endast skrapat på ytan av programmen och rekommenderar att läsaren tar till sig det vi skrivit under resultat och diskussion för att sedan bilda sig en egen uppfattning med hjälp av respektive programs prövotid. Till sist presenteras förslag på vidare studier inom ämnet. / Unmanned Aerial Vehicle (UAV) is a term for a remote controlled airbornevehicle. Since UAV is an acronym for the vehicle itself, Unmanned Aircraft Systems(UAS) has therefore replaced UAV, as it is a concept related to the wholesystem, beside the vehicle it also includes landing, ground station andcommunications link. Within land surveying UAS is a relatively new concept asthe technology has its history mainly associated to the military. The purposeof this study is to analyze and evaluate two software and a cloud service for processingand preparation of orthophotos from data collected with a UAS. The questions tobe answered in this thesis are: Can a cloud service replace an advancedcomputer software for generating orthophotos? Can these produce good resultswithout the help of other GIS software? Which software is the easiest tounderstand and to use and what are the main differences. These questions wereanswered by using collected data, and to get the evaluation fair three datasetshave been created. The software being evaluated are Agisoft PhotoScan andPix4UAV desktop/cloud. The data collection was done by a flight with an octokopterover Fågelmyratippen in Dalarna. The results show that the price quicklybecomes high if only Pix4UAV Cloud is used and that generally PhotoScan ischeaper than Pix4UAV Desktop. The quality report that comes with each projectis easy to understand in PhotoScan but more detailed in Pix4UAV Desktop/Cloud. Despitethe use of same data the results vary when processed, for example the groundresolution. Generally PhotoScan is better at showing the work process. Eachcompany’s support is good and they both have tips and tricks at their websites.On the Agisoft webpage there is a manual available for download and they alsohave a YouTube-channel with instruction videos. The conclusion is that thecloud service is not capable of replacing an advance image processing software.Another conclusion is that for the moment another GIS-program should be used toget the best results. We like to point out that we only scratched the surfaceof the software and we recommend that the reader embrace what we write inresults and discussion to then form their own opinion by using the softwareevaluation period. I the last part we present subjects of further study.

This master thesis deals with the design of the internal wing structure of 3D printed flying wing VUT 714. As a part of this work was performed a load calculation, static strenght test, comparative FEM analysis and subsequently topology optimization was performed. The outcome of this master thesis is a rational design of the wing internal structure based on the load calculation and knowledge of effect of this load on the structure.