The usage of Unmanned Aerial Vehicles (UAV) for several applications has in-creased during the past years. One of the possible applications are aerial imagecapturing for detection and surveillance purposes. In order to make the captur-ing process more efficient, multiple, cameraequipped UAV:s could fly in a for-mation and as a result cover a larger area. To be able to receive several imagesequences and stitch those together, resulting in a panoramavideo, a softwareapplication has been developed and tested for this purpose.All functionality are developed in the language C++ by using the software li-brary OpenCV. All implementations of different techniques and methods hasbeen done as generic as possible to be able to add functionality in the future.Common methods in computervision and object recognition such as SIFT, SURF and RANSAC have been tested. / Användningen av UAV:er för olika tillämpningar har ökat under senare år. Ett avmöjliga användningsområden är flygfotografering och övervakning. För att gö-ra bildupptagningen mer effektiv kan flera UAV:er flyga i formation och på så viskunna fotografera ett avsevärt större område. För att kunna ta in flera bildsekven-ser och foga samman dessa till en panoramavideo, har ett program utvecklats ochtestats för denna uppgift.All funktionalitet för inläsning av bilder och video har utvecklats i C++ medprogrambiblioteket OpenCV. Implementeringen av dessa tekniker och metoderhar gjort så generiskt som möjligt för att det ska vara lättare att lägga till and-ra tekniker och utöka programmets funktioner. Olika tekniker som har testatsinkluderar: SIFT, SURF och RANSAC

A forced landing for an Unmanned Aerial Vehicle (UAV) is required if there is an emergency on board that requires the aircraft to land immediately. Piloted aircraft in the same scenario have a human on board that is able to engage in the complex decision making process involved in the choice of a suitable landing location. If UAVs are to ever fly routinely in civilian airspace, then it is argued that the problem of finding a safe landing location for a forced landing is an important unresolved problem that must be addressed. This thesis presents the results of an investigation into the feasibility of using machine vision techniques to locate candidate landing sites for an autonomous UAV forced landing. The approach taken involves the segmentation of the image into areas that are large enough and free of obstacles; classification of the surface types of these areas; incorporating slope information from readily available digital terrain databases; and finally fusing these maps together using a high level set of simple linguistic fuzzy rules to create a final candidate landing site map. All techniques were evaluated on actual flight data collected from a Cessna 172 flying in South East Queensland. It was shown that the use of existing segmentation approaches from the literature did not provide the outputs required for this problem in the airborne images encountered in the gathered dataset. A simple method was then developed and tested that provided suitably sized landing areas that were free of obstacles and large enough to land. The advantage of this novel approach was that these areas could be extracted from the image directly without solving the difficult task of segmenting the entire image into the individual homogenous objects. A number of neural network classification approaches were tested with the surface types of candidate landing site regions extracted from the aerial images. A number of novel techniques were developed through experimentation with the classifiers that greatly improved upon the classification accuracy of the standard approaches considered. These novel techniques included: automatic generation of suitable output subclasses based on generic output classes of the classifier; an optimisation process for generating the best set of input features for the classifier based on an automated analysis of the feature space; the use of a multi-stage classification approach; and the generation of confidence measures based on the outputs of the neural network classifiers. The final classification result of the system performs significantly better than a human test pilot's classification interpretation of the dataset samples. In summary, the algorithms were able to locate candidate landing site areas that were free of obstacles 92.3 ±2.6% (99% confidence in the result) of the time, with free obstacle candidate landing site areas that were large enough to land in missed only 5.3 ±2.2% (99% confidence in the result) of the time. The neural network classification networks developed were able to classify the surface type of the candidate landing site areas to an accuracy of 93.9 ±3.7% (99% confidence in the result) for areas labelled as Very Certain. The overall surface type classification accuracy for the system (includes all candidate landing sites) was 91.95 ±4.2% (99% confidence in the result). These results were considered to be an excellent result as a human test pilot subject was only able to classify the same data set to an accuracy of 77.24 %. The thesis concludes that the techniques developed showed considerable promise and could be used immediately to enhance the safety of UAV operations. Recommendations include the testing of algorithms over a wider range of datasets and improvements to the surface type classification approach that incorporates contextual information in the image to further improve the classification accuracy.

The last decade has seen significant advances in the development of Vertical takeoff and landing (VTOL) unmanned aerial vehicles (UAVs). The emergence of enabling technologies, in addition to the practical usefulness of such systems has driven their development to a point where numerous technology demonstrators and commercial products are now in existence. Of particular interest has been the development of small scale, VTOL UAVs commonly referred to as mini and micro-VTOL UAVs. The versatility and agility of such vehicles offers great potential for the use in clustered, urban environments. Despite recent advancements, the autonomous navigation of VTOL UAVs remains a very challenging research area. The dynamics of VTOL UAVs are heavily nonlinear, underactuated and non-minimum phase. This, coupled with the aggressive maneuvers that such vehicles are expected to execute provides a stimulating problem in dynamic control. This is particularly true in the case of micro-VTOL UAVs. The fast, nonlinear nature of these systems render classical, linear control approaches inadequate. The past twenty years has seen great interest in the development of nonlinear control strategies. This has led to the emergence of a number of standard design tools, most notably feedback linearisation and Lyapunov-based, backstepping approaches. Such design techniques offer a framework for the derivation of model based control laws capable of achieving global stabilisation and trajectory tracking control for heavily nonlinear systems. Recently, there has been significant interest in the application of such nonlinear control paradigms for the stabilisation and control of VTOL UAVs. The aim of this thesis is to further the application and analysis of nonlinear control design techniques for the control of VTOL UAVs. In particular, focus is placed on Lyapunov-based, backstepping-type control approaches. The first half of this thesis investigates Lyapunov-based control strategies that cast the closed-loop VTOL dynamics into a globally stable, cascade structure. This work was directly inspired by, and builds on, a variety of previously published works. Firstly, an alternative design approach to that previously published is presented, resulting in an improved closed-loop dynamic structure. Although inspired by the VTOL system, this idea may be generalised for the control of a broad class of systems, and is presented as such. A singularity issue arising in the cascade control of VTOL vehicles is then investigated, and a novel approach to overcome this issue is formulated. The second half of this thesis is dedicated to the trajectory tracking control of VTOL UAVs at velocities where the influence of aerodynamics is significant. In general, the aerodynamic models of VTOL UAVs are heavily nonlinear and poorly known. The use of such models in a backstepping framework that uses explicit differentiation of these models for dynamic inversion is questioned, due to the potential sensitivity of such nonlinear models. Consequently, an alternative approach utilising coupled filters to avoid such sensitivity issues is proposed. All control designs formulated in this thesis are accompanied by proofs guaranteeing their global stability, and numerical simulations demonstrating their time domain response characteristics. / http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1298413 / Thesis (Ph.D.) -- University of Adelaide, School of Mechanical Engineering, 2007

The aim of this thesis is to improve understanding of the effects of rotor-rotor interference on small scale rotor systems used on Micro Air Vehicles (MAVs). Previous research on rotor-rotor interactions has focused primarily on planar co-axial and tandem rotors of large scale rotorcraft. The work presented is distinct from prior research not only in its consideration of non-planar rotor systems, but also because of the lower Reynolds numbers and the use of fixed-pitch variable-speed propulsion systems. A design for a novel adjustable rotor interaction test-rig is presented along with a methodology for acquiring accurate and repeatable steady state performance data for two interacting rotor systems. Two six-axis force balances are used to acquire instantaneous and time averaged force and torque data and PIV is used to derive instantaneous and time-averaged flow field data for single and interacting rotor cases. The resulting performance and flow field data represents a unique dataset that can be used in the analysis of small scale rotor interactions, and in the validation of CFD investigations. Results show that for disc angles of between 180 degrees and 90 degrees interactions between rotors are negligible. As the disc angle is reduced from the orthogonal case to the coaxial case interactions significantly effect thrust, pitching moment, and efficiency. It is recommended that in the design of non-planar multirotor vehicles disc angles greater than 75 degrees are utilised to avoid the strong rotor-rotor interactions seen at lower disc angles. A review of existing and future non-planar multirotor concepts shows that the majority avoid significant rotor interactions by virtue of large disc angles.

Small Unmanned Aerial Vehicles (UAVs) have seen continual growth in both research and commercial applications. Attractive features such as their small size, light weight and low cost are a strong driver of this growth. However, these factors also bring about some drawbacks. The light weight and small size means that small UAVs are far more susceptible to performance degradation from factors such as wind gusts. Due to the generally low cost, available sensors are somewhat limited in both quality and available measurements. For example, it is very unlikely that angle of attack is sensed by a small UAV. These aircraft are usually constructed by the end user, so a tangible amount of variation will exist between different aircraft of the same type. Depending on application, additional variation between flights from factors such as battery placement or additional sensors may exist. This makes the application of optimal model based control methods difficult. Research literature on the topic of small UAV control is very rich in regard to high level control, such as path planning in wind. A common assumption in such literature is the existence of a low level control method which is able to track demanded aircraft attitudes to complete a task. Design of such controllers in the presence of significant wind or modelling errors (factors collectively addressed as lumped disturbances herein) is rarely considered. Disturbance Observer Based Control (DOBC) is a means of improving the robustness of a baseline feedback control scheme in the presence of lumped disturbances. The method allows for the rejection of the influence of unmeasurable disturbances much more quickly than traditional integral control, while also enabling recovery of nominal feedback con- trol performance. The separation principle of DOBC allows for the design of a nominal feedback controller, which does not need to be robust against disturbances. A DOBC augmentation can then be applied to ensure this nominal performance is maintained even in the presence of disturbances. This method offers highly attractive properties for control design, and has seen a large rise in popularity in recent years. Current literature on this subject is very often conducted purely in simulation. Ad- ditionally, very advanced versions of DOBC control are now being researched. To make the method attractive to small UAV operators, it would be beneficial if a simple DOBC design could be used to realise the benefits of this method, as it would be more accessible and applicable by many. This thesis investigates the application of a linear state space disturbance observer to low level flight control of a small UAV, along with developments of the method needed to achieve good performance in flight testing. Had this work been conducted purely in simulation, it is likely many of the difficulties encountered would not have been addressed or discovered. This thesis presents four main contributions. An anti-windup method has been devel- oped which is able to alleviate the effect of control saturation on the disturbance observer dynamics. An observer is designed which explicitly considers actuator dynamics. This development was shown to enable faster observer estimation dynamics, yielding better disturbance rejection performance. During initial flight testing, a significant aeroelastic oscillation mode was discovered. This issue was studied in detail theoretically, with a pro- posed solution developed and applied. The solution was able to fully alleviate the effect in flight. Finally, design and development of an over-actuated DOBC method is presented. A method for design of DOBC for over actuated systems was developed and studied. The majority of results in this thesis are demonstrated with flight test data.

The rapid development of Machine-Type-Communication (MTC) has brought big challenges to cellular networks such as super-dense devices and high-shadowing channels which may substantially decrease the spectrum efficiency and increase devices' power consumption. It is pressing to improve the transmission efficiency for MTC due to the limited wireless spectrum. Lower efficiency may also lead to longer transmission time and more energy consumption which conflict with MTC's requirement of lower power consumption. In order to address the above issues, we propose to apply Network Coding (NC) and Device-to-Device (D2D) communications to MTC devices. Our approach introduces an additional delay for local packet exchange, which is acceptable given that MTC traffic typically has the feature of delay tolerance to certain degree. The benefit of the proposed approach is that the cellular transmissions are no longer user-specific, and thus an additional multi-user diversity gain is achieved. The cellular transmission efficiency will also be increased. How to apply the proposed approach for both downlink and uplink has been studied. For the downlink, in addition to the reduction of cellular resource consumption, the MTC devices' feedback load can also be significantly reduced because the cellular transmissions are not sensitive to user-specific errors. In the uplink, besides the enhanced transmission efficiency for full-buffer traffic, an additional small-data aggregation gain is achieved for MTC small-data traffic. Theoretical performance analyses for both downlink and uplink and the corresponding numerical evaluations are given. Though the proposed NC and D2D approach can improve the transmission efficiency by exploring multi-user diversity gain, poor-quality MTC channels still exist which affect system performance. When the whole group MTC devices in an area experience high shadowing and penetration loss, we have to increase either the resource consumption or the transmitting power to overcome the poor-quality channels. The existing small-cell solution can improve the MTC channel quality, but MTC's unique traffic characteristics and quality of service requirements, as well as other practical issues, make the small-cell deployment unprofi table. Therefore, we propose a solution using Floating Relay (FR) given the mature technologies of Unmanned Aerial Vehicle (UAV). We rst target on the high-shadowing channels of the MTC devices and introduce the FR into the cellular system to improve the transmission efficiency and maximize the system capacity. An optimization problem, given the capacity limit of the FR's back-haul link and the maximum transmission power of each user, is formulated and then theoretically solved. An effective on-line fight path planning algorithm is also proposed. Then, we extend the FR concept to a bigger picture and propose the UAV-assisted heterogeneous cellular solution. Detailed system design and comprehensive analyses on FR-cells deployment including frequency reuse, interference, backhaul resource allocation, and coverage are given. For UAV assisted networking systems, mobility and topology play important roles. How to dispatch a UAV to the optimal location in a mesh network to enhance the coverage and service of the existing network is a critical issue. Given the topology of existing service nodes, a new supplementary UAV can be sent to improve the quality of service especially for the users with poor-quality channels. The location of a newly added UAV is optimized to improve the service quality to the worst point. In summary, we propose two means to improve the transmission efficiency for MTC in this thesis work. The NC and D2D approach can be used when some of the MTC devices have chances to experience better channels because of the fast fading and uneven shadowing. Otherwise, the FR can be applied to proactively improve the channel quality for MTC. The NC and D2D approach sticks to the latest standard in the cellular system and thus provides a down-to-earth and backward-compatible MTC solution for 5G cellular system. The UAV-assisted heterogeneous cellular solution and UAV mesh networks can enable mobile Internet and ultra-reliable low latency communications, respectively. These solutions together effectively and efficiently support MTC, which is key to future proliferation of Internet of Things / Graduate

Throughout the past 20 years use of unmanned aerial vehicles, or drones, have increased immensely. The US has since the beginning of the 21st century employed armed drones as their weapon of choice in the war on terror. As a result, a lot of questions concerning the ethical aspects of the drone has been raised. This thesis aims to clarify whether these aspects can be linked to the drone’s characteristics as unmanned, or if they’re rather connected to the tactics utilised by the Americans. Furthermore, the conclusions drawn from this can help countries like Sweden, understand if the use of armed drones would result in the same split in public opinion as in the US. The analysis shows, through a survey of US media, that the ethical debate mainly focuses on aspects linked to how drones are used by the US. Consequently, it seems plausible that the mere acquisition of armed drones by Sweden is not enough to launch a major debate about the ethics of unmanned aircraft.

Approved for public release; distribution is unlimited. / An autonomously guided rocket-powered delivery vehicle has been under development at the Naval Postgraduate School. Designed to eventually counter UAV swarm attacks, the vehicle made advances toward reaching a target in the sky. These advances reduced the time needed to launch, modify, and relaunch the rocket, while adding capabilities such as data transfer along the vehicle axis and the rapid download of flight data. Improving the vehicle included reconfiguring the guidance, navigation, and control (GNC) strategy. Advancements included the design, implementation, and evaluation of electronic servo control, actuating fins, and the mechanical coupling design. The forward compartment in the vehicle’s nose cone was structurally modified for the GNC equipment and to support electronics under high-g launch conditions. Modifications included innovative designs for managing heat transfer requirements. Using off-the-shelf subsystem components kept the advancements fiscally mindful. After implementing the design features, two final test launches were performed: one demonstrated a control spin rate of 8.5 rad/sec; the other showed the vehicle’s ability to execute pitch maneuvers on a single axis. The test results can be used to improve the GNC software and servo control parameters. Continued development will allow the system to become a viable option for countering UAV swarms. / Ensign, United States Navy

This thesis is concerned with the challenging tasks of automatic and real-time vehicle detection and tracking from aerial video. The aim of this thesis is to build an automatic system that can accurately localise any vehicles that appear in aerial video frames and track the target vehicles with trackers. Vehicle detection and tracking have many applications and this has been an active area of research during recent years; however, it is still a challenge to deal with certain realistic environments. This thesis develops vehicle detection and tracking algorithms which enhance the robustness of detection and tracking beyond the existing approaches. The basis of the vehicle detection system proposed in this thesis has different object categorisation approaches, with colour and texture features in both point and area template forms. The thesis also proposes a novel Self-Learning Tracking and Detection approach, which is an extension to the existing Tracking Learning Detection (TLD) algorithm. There are a number of challenges in vehicle detection and tracking. The most difficult challenge of detection is distinguishing and clustering the target vehicle from the background objects and noises. Under certain conditions, the images captured from Unmanned Aerial Vehicles (UAVs) are also blurred; for example, turbulence may make the vehicle shake during flight. This thesis tackles these challenges by applying integrated multiple feature descriptors for real-time processing. In this thesis, three vehicle detection approaches are proposed: the HSV-GLCM feature approach, the ISM-SIFT feature approach and the FAST-HoG approach. The general vehicle detection approaches used have highly flexible implicit shape representations. They are based on training samples in both positive and negative sets and use updated classifiers to distinguish the targets. It has been found that the detection results attained by using HSV-GLCM texture features can be affected by blurring problems; the proposed detection algorithms can further segment the edges of the vehicles from the background. Using the point descriptor feature can solve the blurring problem, however, the large amount of information contained in point descriptors can lead to processing times that are too long for real-time applications. So the FAST-HoG approach combining the point feature and the shape feature is proposed. This new approach is able to speed up the process that attains the real-time performance. Finally, a detection approach using HoG with the FAST feature is also proposed. The HoG approach is widely used in object recognition, as it has a strong ability to represent the shape vector of the object. However, the original HoG feature is sensitive to the orientation of the target; this method improves the algorithm by inserting the direction vectors of the targets. For the tracking process, a novel tracking approach was proposed, an extension of the TLD algorithm, in order to track multiple targets. The extended approach upgrades the original system, which can only track a single target, which must be selected before the detection and tracking process. The greatest challenge to vehicle tracking is long-term tracking. The target object can change its appearance during the process and illumination and scale changes can also occur. The original TLD feature assumed that tracking can make errors during the tracking process, and the accumulation of these errors could cause tracking failure, so the original TLD proposed using a learning approach in between the tracking and the detection by adding a pair of inspectors (positive and negative) to constantly estimate errors. This thesis extends the TLD approach with a new detection method in order to achieve multiple-target tracking. A Forward and Backward Tracking approach has been proposed to eliminate tracking errors and other problems such as occlusion. The main purpose of the proposed tracking system is to learn the features of the targets during tracking and re-train the detection classifier for further processes. This thesis puts particular emphasis on vehicle detection and tracking in different extreme scenarios such as crowed highway vehicle detection, blurred images and changes in the appearance of the targets. Compared with currently existing detection and tracking approaches, the proposed approaches demonstrate a robust increase in accuracy in each scenario.

The field of UAV systems is an active research area with potential for development and enhancement in various perspectives. This thesis investigates different issues related to the design, operation and control of UAV systems with a focus on the application side of each proposed solution where the implementation side and applicability of the proposed solutions are always considered with high priority. The thesis discusses unmodeled actuator dynamics and their effect on UAV systems when using feedback linearisation to linearize nonlinear models of UAVs. The analysis shows potential risk when implementing feedback linearisation and neglecting actuator dynamics even for first order actuator system. A solution algorithm of two stage feedback linearisation is proposed to handle actuator dynamics and linearize the main dynamics of the system. In the field of design and operation of UAVs, this thesis proposes a systematic design procedure for electric propulsion systems that are widely used in UAVs. The design procedure guides the designer step by step to achieve minimum propulsion system weight or maximum flight time or a trade off between the two factors from the supplied solution sets. On the navigation side, the thesis proposes a new indoor navigation system that is easy to implement and less costly compared with other indoor navigation systems. The proposed system can be classified under computer-vision based navigation systems, however, it needs less information and less computational capacity. The thesis also contributes to the structure design of UAV systems by producing a novel tri-rotor UAV platform. The proposed UAV is novel in structure and design and has a centralized control system that stabilizes and tracks both rotational and transitional motion of the vehicle simultaneously.