Les mini-drones à propulsion électrique sont susceptibles d’avoir une endurance inférieure à celle de drones plus grands.L’exploitation des interactions aérodynamiques, inspirée par les oiseaux migratoires, ainsi que le ravitaillement en vol , sont des approches prometteuses pour améliorer l’endurance des mini-drones. La commande par modes glissants d’ordre supérieur en temps continu (CTHOSM) a été considérée comme un candidat prometteur à ce problème ouvert difficile et a été appliquée avec succès à des modèles cinématiques simples. Dans nos travaux, nous étudions les implications de la présence de la dynamique de la boucle interne et de l’implémentation en temps discret à des taux d’échantillonnage modérés et constatons alors que l’application de la commande CTHOSM devient impossible. Nous proposons donc un schéma de guidage prédictif discret par modes glissants pour approximer les performances de la commande CTHOSM pour une dynamique réaliste du drone. On propose également un problème de référence accessible pour d'autres chercheurs. Les algorithmes de localisation probabilistes existants ne permettent pas la caractérisation de régions de confiance garanties de la position des autres membres de la formation. Dans ce contexte, nous proposons un nouveau filtre ensembliste caractérisant de telles régions de confiance sous forme ellipsoïdale. Nos premières évaluations ont montré que les efforts de calcul induits par cette mise en œuvre restent parfaitement compatibles avec les contraintes des systèmes avioniques des petits drones. / Small, electrically driven unmanned aircraft are likely to suffer from inferior endurance compared to their larger counterparts. Upwash exploitation by tight formation flight, as well as aerial recharging are the most promising control-driven approaches to mitigate this disadvantage. Continuous time higher order sliding mode control (CTHOSM) has been considered as a candidate for this challenging open problem and was successfully applied to simple kinematic models in simulation, where excellent relative position tracking performance can be demonstrated. In this work we study the implications of the presence of inner loop dynamics and discrete implementation at moderate sampling rates and we find that it precludes the application of CTHOSM control to fixed-wing UAS. We propose a predictive discrete sliding mode guidance scheme to approximate the performance of CTHOSM control assuming realistic fixed-wing UAS dynamics. We show that the proposed guidance scheme in combination with inner load factor tracking loops and a disturbance observer allows for relative position tracking performance compatible with the requirements of upwash exploitation. We propose as well an openly accessible benchmark problem. Existing probabilistic localization algorithms cannot provide guaranteed confidence regions of the relative position between UAS. We present a set membership filter that provides ellipsoidal regions guaranteed to contain the relative positions of the other UAS. It is compatible with the hardware constraints of small low-cost UAS. Simulations suggest computational efforts compatible with the computational resources typically available onboard small UAS.

近年來多軸空拍機快速發展與普及，可預見其未來將成為新一代的輔助攝影工具。空拍機打破距離、角度的限制，讓時下流行的自拍（Selfie）照片相較於以往以手臂或自拍棒輔助的方式，拍出更具特色及多樣性的構圖。然而當前空拍機操作方式複雜，使用者需花費一定的練習時間才能熟練地控制空拍機至預定位置進行拍攝。 本研究針對空拍機的使用情境進行觀察，歸納傳統操作方式造成的問題，再依不同互動控制方法討論過往研究提出的解決方法之優勢與限制，並依據自拍行為之心智模型及過往研究經驗，設計兩階段互動流程：相機定位階段（Positioning）及鏡頭構圖階段（Framing），並在各階段中分別提出直接指向（Direct Pointing）、移動微調（Fine Tuning）及構圖調整（Framing）三種有別於傳統類比搖桿操控之互動模式。 本研究另設計一俱觸覺控制回饋、可單手操作之實體自拍遙控器，搭配前述互動設計實作兼具自動化移動及以使用者為中心進行操作之空拍機自拍互動控制系統原型Skyfie，並於戶外環境設計實驗場域進行使用者測試，測試使用者指揮空拍機至指定位置拍照之操作，評估互動流程中各操作方法之優劣。測試結果顯示Skyfie 互動控制系統相較於傳統的類比搖桿操控方式更易於學習及使用，且符合使用者對空間的認知，並依照回饋意見進行互動模式修正，以達成對空拍機初學者而言也簡單易學的互動操作方式。 / As personal drones become more popular, we can envision a future where flying selfie bots are always with us. Drones break the limits of distance and angle, providing more diversified composition than taking selfie with arms or a selfie stick. However, users have to be very skillful to pilot the drone and are not easy to take aerial selfies by state-of-the-art methods. Based on user observation and the mindset of selfie taking, we summarize the problems caused by the traditional control methods and generalize the interaction flow of selfie taking into two stages: Positioning and Framing stage. In each stage, we present new interaction techniques including a direct pointing technique, fine-tuning technique, and a touch manipulation for framing. We also present a selfie remote controller designed for single-hand operation to collocates with the interaction techniques, and implement a proof-of-concept Skyfie system for an outdoor user testing. The result shows users felt intuitive and expressed enthusiasm to take aerial selfies with our techniques. Finally, we discuss the insights from the evaluation and conclude with future directions.

La planification HTN (Hierarchical Task Network, ou Réseau Hiérarchique de Tâches) est une approche très souvent utilisée pour produire des séquences de tâches servant à contrôler des systèmes intelligents. Cette thèse présente le planificateur HATP (Hierarchical Agent-base Task Planner, ou Planificateur Hiérarchique centré Agent) qui étend la planification HTN classique en enrichissant la représentation des domaines et leur sémantique afin d'être plus adaptées à la robotique, tout en offrant aussi une prise en compte des humains. Quand on souhaite générer un plan pour des robots tout en prenant en compte les humains, il apparaît que les problèmes sont complexes et fortement interdépendants. Afin de faire face à cette complexité, nous avons intégré à HATP un planificateur géométrique apte à déduire l'effet réel des actions sur l'environnement et ainsi permettre de considérer la visibilité et l'accessibilité des éléments. Cette thèse se concentre sur l'intégration de ces deux planificateurs de nature différente et étudie comment par leur combinaison ils permettent de résoudre de nouvelles classes de problèmes de planification pour la robotique. / Hierarchical Task Network (HTN) planning is a popular approach to build task plans to control intelligent systems. This thesis presents the HATP (Hierarchical Agent-based Task Planner) planning framework which extends the traditional HTN planning domain representation and semantics by making them more suitable for roboticists, and by offering human-awareness capabilities. When computing human-aware robot plans, it appears that the problems are very complex and highly intricate. To deal with this complexity we have integrated a geometric planner to reason about the actual impact of actions on the environment and allow to take into account the affordances (reachability, visibility). This thesis presents in detail this integration between two heterogeneous planning layers and explores how they can be combined to solve new classes of robotic planning problems

As science and technology continue to advance, innovative developments in transportation can enhance product safety and security for the benefit and welfare of society. The federal government requires every commercial truck to be inspected before each trip. This pre-trip inspection ensures the safe mechanical condition of each vehicle before it is used. An Unmanned Aerial Vehicle (UAV) could be used to provide an automated inspection, thus reducing driver workload, inspection costs and time while increasing inspection accuracy. This thesis develops a primary component of the algorithm that is required to implement UAV pre-trip inspections for commercial trucks using an android-based application. Specifically, this thesis provides foundational work of providing stable height control in an outdoor environment using a laser sensor and an android flight control application that includes take-off, landing, throttle control, and real-time video transmission. The height algorithm developed is the core of this thesis project. Phantom 2 Vision+ uses a pressure sensor to calculate the altitude of the drone for height stabilization. However, these altitude readings do not provide the precision required for this project. Rather, the goal of autonomously controlling height with great precision necessitated the use of a laser rangefinder sensor in the development of the height control algorithm. Another major contribution from this thesis research is to extend the limited capabilities of the DJI software development kit in order to provide more sophisticated control goals without modifying the drone dynamics. The results of this project are also directly applicable to a number of additional uses of drones in the transportation industry.

In recent years, companies such as Amazon and DHL have started to develop alternative last-mile delivery options because of the growing trend in e-commerce. Last-mile deliveries refer to the terminal to customer deliveries and have become increasingly challenging to manage, as traffic congestions in cities has risen. Therefore, companies have begun searching for alternative delivery methods, such as autonomous drones or unmanned aerial vehicles (UAV). The purpose of this thesis is to investigate drones in the current literature on last-mile delivery. First, by conducting a systematic literature review on the current literature available in two major databases. Second, by analyzing the collected literature sample and presenting it through a descriptive analysis focusing on bibliometrics and a thematic analysis that identifies emerging themes for drivers for drones’ application in last-mile delivery, barriers for drones’ application in last-mile delivery, and approaches for implementing drones in last-mile delivery.                        Drones’ application in last-mile delivery has the potential to reduce cost, decrease delivery time, reduce emissions, and reduce energy consumption. However, barriers that hinder the implementation, such as governmental regulations, need further consideration for the implementation to occur. Some evident gaps can be found related to the barriers. Both managers and researchers might find the thesis useful, as it provides a holistic view on the subject, with theoretical, managerial, and societal implications being presented.

Denna rapport handlar om sju studenters kandidatarbete som utfördes i kursen TDDD96 - Kandidatprojekt i programvaruutveckling vid Linköpings universitet under våren 2020. Målet med projektet var att utveckla en webbapplikation som visar en karta, där användaren kan styra drönare genom att specificera ett område på denna karta. Drönarna tar flygfoton över området som sedan visas på kartan. Resultatet blev en fungerande produkt för demonstrationssyften, en teknisk beskrivning av produkten och en användarmanual. Rapporten innehåller även sju individuella delar som ger en fördjupning inom olika delområden av projektet.

This thesis focuses on vision-based detection and observation of small, slow-moving targets using a gimballed fixed-wing unmanned aircraft system (UAS). Generally, visual tracking algorithms are tuned to detect motion of relatively large objects in the scene with noticeably significant motion; therefore, applications such as high-altitude visual searches for human motion often ignore target motion as noise. Furthermore, after a target is identified, arbitrary maneuvers for transitioning to overhead orbits for better observation may result in temporary or permanent loss of target visibility. We present guidelines for tuning parameters of the Visual Multiple Target Tracking (Visual MTT) algorithm to enhance its detection capabilities for very small, slow-moving targets in high-resolution images. We show that the tuning approach is able to detect walking motion of a human described by 10-15 pixels from high altitudes. An algorithm is then presented for defining rotational bounds on the controllable degrees of freedom of an aircraft and gimballed camera system for maintaining visibility of a known ground target. Critical rotations associated with the fastest loss or acquisition of target visibility are also defined. The accuracy of these bounds are demonstrated in simulation and simple applications of the algorithm are described for UAS. We also present a path planning and control framework for defining and following both dynamically and visually feasibly transition trajectories from an arbitrary point to an orbit over a known target for further observation. We demonstrate the effectiveness of this framework in maintaining constant target visibility while transitioning to the intended orbit as well as in transitioning to a lower altitude orbit for more detailed visual analysis of the intended target.

The Wright fishhook cactus (Sclerocactus wrightiae L.D. Benson) is an endangered cactus species endemic to south-central Utah. Since its listing in 1979 by the U.S. Fish and Wildife Service, the potential impacts of soil disturbance by cattle have become a central focus of management policies and monitoring efforts. However, little to no empirical data has been collected to substantiate the hypothesis that soil disturbance by cattle has direct or indirect negative effects on Wright fishhook cactus growth or reproduction. Over the years, the Bureau of Land Mangement (BLM) and Capitol Reef National Park (CRNP) have invested significant resources documenting cactus populations including several attributes of individual cacti: GPS location, diameter, number of flowers, fruits, or buds, number of stems, and the presence or absence of a cow track within 15 cm of the cactus. While these efforts have been commendable, due to the defining phenological characteristics of this species (flower and filament color) and its short flowering period (April-May) it remains difficult to study and much basic biological information including a range wide population estimate and defined critical habitat remain unknown. Our research had two primary objectives, 1) evalutate the effects of soil disturbance by cattle on reproduction and diameter of the Wright fishhook cactus (Chapters 1 and 2), and 2) explore the use of drones and GIS to define critical habitat and obtain an accurate range wide population estimate (Chapters 3 and 4). In Chapter 1, we analyzed cactus attribute data collected by the BLM at 30 macro-plots representing different levels of soil related cattle disturbance (high, moderate, and low) from 2011-2017. We found no significant association between level of cattle disturbance and flower density or cactus diameter. We did find a significant negative association between flower frequency and increased disturbance. In Chapter 2, we conducted an experimental study where tracks were simulated within 15 cm of cacti at various levels (Ctrl, 1-Track, 2-Track, 4-Tracks, and 4-Tracks Doubled). No significant association was observed between the number of tracks and response in diameter, flower production, fruit production, or seed set. In Chapter 3, we conducted drone flights over 14 macro-plots at three different altitudes above ground level (10 m, 15 m, and 20 m) and found that while the 10 m flights provided the best remotely sensed survey results, drones are not a suitable replacement for ground censuses. In Chapter 4, we used Resource Selection Function to define critical habitat for the Wright fishhook cactus. Our modeling suggests that geology, elevation, and slope are significant factors in defining cactus habitat. Based on the results of our research we conclude that soil disturbance by cattle may not have a significant influence on Wright fishhook cactus populations or dynamics, and that accurate range wide population estimates may be best obtained through ground surveys within the predicted critical habitat.

The Wright fishhook cactus (Sclerocactus wrightiae L.D. Benson) is an endangered cactus species endemic to south-central Utah. Since its listing in 1979 by the U.S. Fish and Wildife Service, the potential impacts of soil disturbance by cattle have become a central focus of management policies and monitoring efforts. However, little to no empirical data has been collected to substantiate the hypothesis that soil disturbance by cattle has direct or indirect negative effects on Wright fishhook cactus growth or reproduction. Over the years, the Bureau of Land Mangement (BLM) and Capitol Reef National Park (CRNP) have invested significant resources documenting cactus populations including several attributes of individual cacti: GPS location, diameter, number of flowers, fruits, or buds, number of stems, and the presence or absence of a cow track within 15 cm of the cactus. While these efforts have been commendable, due to the defining phenological characteristics of this species (flower and filament color) and its short flowering period (April-May) it remains difficult to study and much basic biological information including a range wide population estimate and defined critical habitat remain unknown. Our research had two primary objectives, 1) evalutate the effects of soil disturbance by cattle on reproduction and diameter of the Wright fishhook cactus (Chapters 1 and 2), and 2) explore the use of drones and GIS to define critical habitat and obtain an accurate range wide population estimate (Chapters 3 and 4). In Chapter 1, we analyzed cactus attribute data collected by the BLM at 30 macro-plots representing different levels of soil related cattle disturbance (high, moderate, and low) from 2011-2017. We found no significant association between level of cattle disturbance and flower density or cactus diameter. We did find a significant negative association between flower frequency and increased disturbance. In Chapter 2, we conducted an experimental study where tracks were simulated within 15 cm of cacti at various levels (Ctrl, 1-Track, 2-Track, 4-Tracks, and 4-Tracks Doubled). No significant association was observed between the number of tracks and response in diameter, flower production, fruit production, or seed set. In Chapter 3, we conducted drone flights over 14 macro-plots at three different altitudes above ground level (10 m, 15 m, and 20 m) and found that while the 10 m flights provided the best remotely sensed survey results, drones are not a suitable replacement for ground censuses. In Chapter 4, we used Resource Selection Function to define critical habitat for the Wright fishhook cactus. Our modeling suggests that geology, elevation, and slope are significant factors in defining cactus habitat. Based on the results of our research we conclude that soil disturbance by cattle may not have a significant influence on Wright fishhook cactus populations or dynamics, and that accurate range wide population estimates may be best obtained through ground surveys within the predicted critical habitat.

The use of drones in conflicts is under development and is increasing rapidly. The first real drone warfare was seen in Pakistan in the War on Terrorism in 2004. Between officials and scientists, there are divided opinions about whether the drone strikes can increase terrorism or cause the collapse of organizations. The purpose of this thesis is to examine how the effectiveness of drones, in terms of precision and lethality, can be seen to escalate or de-escalate a conflict. The aim is to explore whether two existing theories about air power can explain the case of killing the Iranian general, Qasem Soleimani, in 2020. As a qualitative case study, the approach results in favor of Pape´s theory over Warden´s. Pape says that decapitation is more likely to escalate a conflict, which confirms this case. Other explanations in his theory are the absence of unexpected political effects and the overthrow of government. Though Wardens theory shows some aspects of de-escalation, the conclusion is that Soleimani was not the center of gravity that was meant to cause system collapse. Further research is required since the findings cannot identify any escalating or de-escalating effects within the Quds Force.