The Micro Aerial Vehicless (MAVs) are gaining attention in numerous applications asthese platforms are cheap and can do complex maneuvers. Moreover, most of the commer-cially available MAVs are equipped with a mono-camera. Currently, there is an increasinginterest to deploy autonomous mono-camera MAVs with obstacle avoidance capabilitiesin various complex application areas. Some of the application areas have moving obstaclesas well as stationary, which makes it more challenging for collision avoidance schemes.This master thesis set out to investigate the possibility to avoid moving and station-ary obstacles with a single camera as the only sensor gathering information from thesurrounding environment.One concept to perform autonomous obstacle avoidance is to predict the time near-collision based on a Convolution Neural Network (CNN) architecture that uses the videofeed from a mono-camera. In this way, the heading of the MAV is regulated to maximizethe time to a collision, resulting in the avoidance maneuver. Moreover, another interestingperspective is when due to multiple dynamic obstacles in the environment there aremultiple time predictions for different parts of the Field of View (FoV). The method ismaximizing time to a collision by choosing the part with the largest time to collision.However, this is a complicated task and this thesis provides an overview of it whilediscussing the challenges and possible future directions. One of the main reason was thatthe available data set was not reliable and was not provide enough information for theCNN to produce any acceptable predictions.Moreover, this thesis looks into another approach for avoiding collisions, using objectdetection method You Only Lock Once (YOLO) with the mono-camera video feed. YOLOis a state-of-the-art network that can detect objects and produce bounding boxes in real-time. Because of YOLOs high success rate and speed were it chosen to be used in thisthesis. When YOLO detects an obstacle it is telling where in the image the object is,the obstacle pixel coordinates. By utilizing the images FoV and trigonometry can pixelcoordinates be transformed to an angle, assuming the lens does not distort the image.This position information can then be used to avoid obstacles. The method is evaluated insimulation environment Gazebo and experimental verification with commercial availableMAV Parrot Bebop 2. While the obtained results show the efficiency of the method. To bemore specific, the proposed method is capable to avoid dynamic and stationary obstacles.Future works will be the evaluation of this method in more complex environments with multiple dynamic obstacles for autonomous navigation of a team of MAVs. A video ofthe experiments can be viewed at:https://youtu.be/g_zL6eVqgVM.

No description available.

This thesis presents the design of an autopilot capable of flying small unmanned aerial vehicles with wingspans less then 21 inches. The autopilot is extremely small and lightweight allowing it to fit in aircraft of this size. The autopilot features an advanced, highly autonomous flight control system with auto-launch and auto-landing algorithms. These features allow the autopilot to be operated by a wide spectrum of skilled and unskilled users. Innovative control techniques implemented in software, coupled with light weight, robust, and inexpensive hardware components were used in the design of the autopilot.

Though still in relative infancy, the field of flapping flight has potential to have a far-reaching impact on human life. Nature presents a myriad of examples of successful uses of this locomotion. Human efforts in flapping flight have seen substantial improvement in recent times. Wing kinematics are a key aspect of this study. This study summarizes previous wing trajectory generators and presents a new trajectory generation method built upon previous methods. This includes a novel means of commanding unequal half-stroke durations subject to robotic trajectory continuity requirements. Additionally, previous optimization methods are improved upon. Experimental optimization is performed using the new trajectory generation method and a more traditional means. Methods for quantifying and compensating for sensor time-dependence are also discussed. Results show that the Polar Fourier Series trajectory generator advanced rapidly through the optimization process, especially during the initial phase of experimentation. The Modified Berman and Wang trajectory generator moved through the design space more slowly due to the increased number of kinematic parameters. When optimizing lift only, the trajectory generators produced similar results and kinematic forms. The findings suggest that the objective statement should be modified to reward efficiency while maintaining a certain amount of lift. It is expected that the difference between the capabilities of the two trajectory generators will become more apparent under such conditions.

Coalitions are collections of agents that join together to solve a common problem that either cannot be solved individually or can be solved more efficiently as a group. Each individual agent has capabilities that can benefit the group when working together as a coalition. Typically, individual capabilities are joined together in an additive way when forming a coalition. This work will introduce a new operator that is used when combining capabilities, and suggest that the behavior of the operator is contextual, depending on the nature of the capability itself. This work considers six different capabilities of Unmanned Air Vehicles (UAV) and determines the nature of the new operator in the context of each capability as coalitions (squadrons) of UAVs are formed. Coalitions are formed using three different search algorithms, both with and without heuristics: Depth-First, Depth-First Iterative Deepening, and Genetic Algorithm (GA). The effectiveness of each algorithm is evaluated. Multi agent-based UAV simulation software was developed and used to test the ideas presented. In addition to coalition formation, the software aims to address additional multi-agent issues such as agent identity, mutability, and communication as applied to UAV systems, in a realistic simulated environment. Social potential fields provide a means of modeling a clustering attractive force at the same time as a collision-avoiding repulsive force, and are used by the simulation to maintain aircraft position relative to other UAVs.

Unmanned Aerial Vehicles (UAVs) provides a platform for many automated tasks and with an ever increasing advances in computing, these tasks can be more complex. The use of UAVs is expanded in this thesis with the goal of Search and Rescue (SAR), where a UAV can assist fast responders to search for a lost person and relay possible search areas back to SAR teams. To identify a person from an aerial perspective, low-level Histogram of Oriented Gradients (HOG) feature descriptors are used over a segmented region, provided from thermal data, to increase classification speed. This thesis also introduces a dataset to support a Bird’s-Eye-View (BEV) perspective and tests the viability of low level HOG feature descriptors on this dataset. The low-level feature descriptors are known as Boosted Histogram of Oriented Gradients (BHOG) features, which discretizes gradients over varying sized cells and blocks that are trained with a Cascaded Gentle AdaBoost Classifier using our compiled BEV dataset. The classification is supported by multiple sensing modes with color and thermal videos to increase classification speed. The thermal video is segmented to indicate any Region of Interest (ROI) that are mapped to the color video where classification occurs. The ROI decreases classification time needed for the aerial platform by eliminating a per-frame sliding window. Testing reveals that with the use of only color data iv and a classifier trained for a profile of a person, there is an average recall of 78%, while the thermal detection results with an average recall of 76%. However, there is a speed up of 2 with a video of 240x320 resolution. The BEV testing reveals that higher resolutions are favored with a recall rate of 71% using BHOG features, and 92% using Haar-Features. In the lower resolution BEV testing, the recall rates are 42% and 55%, for BHOG and Haar-Features, respectively.

In wireless communications, employing pilot symbols enables to estimate the state of the wireless channel at the expense of decreasing the number of symbols available for transmitting data. Addressing this trade-off is particularly challenging when the channel changes rapidly over time, since channel estimates become obsolete over short transmission periods. This master thesis proposes an analytical model to characterize this trade-off and develops an algorithm to find the near-optimal pilot spacing in terms of the achieved over spectral efficiency. This algorithm is simulated in a cellular system that serves uncrewed aerial vehicles. Numerical results indicate that the altitude of the uncrewed aerial vehicle, the Rician factor, the Doppler frequency, and the number of receive antennas influence the overall spectral efficiency and consequently, pilot spacing should take into account these system parameters. / I trådlös kommunikation möjliggör användning av pilotsymboler att uppskatta tillståndet för den trådlösa kanalen på bekostnad av att minska antalet tillgängliga symboler för att överföra data. Att ta itu med denna avvägning är särskilt utmanande när kanalen ändras snabbt över tiden, eftersom kanaluppskattningar blir föråldrade under korta överföringsperioder. Denna masteruppsats föreslår en analytisk modell för att karakterisera denna avvägning och utvecklar en algoritm för att hitta det närmast optimala pilotavståndet i termer av uppnådd över spektral effektivitet. Denna algoritm simuleras i ett cellulärt system som betjänar obemannade flygfarkoster. Numeriska resultat indikerar att höjden för det obemannade luftfartyget, Rician-faktorn, Dopplerfrekvensen, antalet mottagarantenner påverkar den totala spektrala effektiviteten och följaktligen bör pilotavståndet ta hänsyn till dessa systemparametrar.

No description available.

This thesis aims to analyze how object detectors perform under winter weather conditions, specifically in areas with varying degrees of snow cover. The investigation will evaluate the effectiveness of commonly used object detection methods in identifying vehicles in snowy environments, including YOLO v8, Yolo v5, and Faster R-CNN. Additionally, the study explores the method of labeling vehicle objects within a set of image frames for the purpose of high-quality annotations in terms of correctness, details, and consistency. Training data is the cornerstone upon which the development of machine learning is built. Inaccurate or inconsistent annotations can mislead the model, causing it to learn incorrect patterns and features. Data augmentation techniques like rotation, scaling, or color alteration have been applied to enhance some robustness to recognize objects under different alterations. The study aims to contribute to the field of deep learning by providing valuable insights into the challenges of detecting vehicles in snowy conditions and offering suggestions for improving the accuracy and reliability of object detection systems. Furthermore, the investigation will examine edge devices' real-time tracking and detection capabilities when applied to aerial images under these weather conditions. What drives this research is the need to delve deeper into the research gap concerning vehicle detection using drones, especially in adverse weather conditions. It highlights the scarcity of substantial datasets before Mokayed et al. published the Nordic Vehicle Dataset. Using unmanned aerial vehicles(UAVs) or drones to capture real images in different settings and under various snow cover conditions in the Nordic region contributes to expanding the existing dataset, which has previously been restricted to non-snowy weather conditions. In recent years, the leverage of drones to capture real-time data to optimize intelligent transport systems has seen a surge. The potential of drones in providing an aerial perspective efficiently collecting data over large areas to precisely and timely monitor vehicular movement is an area that is imperative to address. To a greater extent, snowy weather conditions can create an environment of limited visibility, significantly complicating data interpretation and object detection. The emphasis is set on edge devices' real-time tracking and detection capabilities, which in this study introduces the integration of edge computing in drone technologies to explore the speed and efficiency of data processing in such systems.

The use of unmanned aerial vehicles (UAVs) is well established in the military sector with great advantages in modern warfare. The concept of using UAV swarms has been discussed over two decades, but is now seeing its first real system used by the Israel defence forces. There is no exact definition what a swarm is, but it is proposed that it should satisfy the following three requirements. A swarm should have limited human control, the number of agents in a swarm should be at least three and the agents in the swarm should cooperate to perform common tasks. The complexity of controlling multiple autonomous UAVs gives way to the problem of how to take advantage of the operators cognitive and tactical abilities to control a swarm to effectively conduct military reconnaissance missions. The method of using behaviour trees as a control structure was derived from previous work in swarm systems. A behaviour tree is a structured way of organising and prioritising actions of autonomous systems. Behaviour trees are similar to finite state machines (FSMs) with the advantages of being modular, reactive, and with better readability. Three different behaviour trees with increasing complexity was created and simulated in the game engine Unity. A fourth more real life behaviour tree was created and used as a basis for discussing the strength and weaknesses of using behaviour trees against previous work. Using behaviour tree as a unifying structure for creating a swarm that integrates the tactical abilities of an operator with the strength of an autonomous swarm seems promising. The proposed method of using behaviour trees i suggested to be used as a platform for discussing the swarm desired functions and to create a common vision for both operators and engineers how a swarm should function. / Användning av drönare är väletablerad inom det militära och ger stora fördelar i dagens moderna krigsföring. Idén om att använda en svärm av drönare har diskuterats under de senaste två decennierna, men först nu sett sin första riktiga system som använts av Israels försvarsmakt. Det finns ingen exakt definition av vad en svärm är, men det har föreslagits att en svärm ska uppfylla de följande tre kraven. En svärm ska ha begränsad mänsklig interaktion, antalet agenter ska vara minst tre och svärmen ska samarbeta för att lösa gemensamma uppgifter. Svårigheterna med att styra en autonom svärm ger upphov till hur man ska utnyttja en operatörs kognitiva och taktiska förmåga för att styra en autonom drönarsvärm för att effektivt utföra militära spaningsuppdrag. Utifrån tidigare arbete inom styrning av svärmar verkade beteende träd som en lovande metod. Beteendeträd är ett strukturerat sätt att organisera och prioritera beteenden för ett autonomt system. Beteendeträd har många likheter med ändliga tillståndsmaskiner, men fördelarna att vara modulära, responsiva och mer lättläsliga. Tre olika beteendeträd med ökande komplexitet skapades och deras funktionalitet simulerades i Unity. Ett fjärde mer verklighetstroget beteendeträd skapades och användes som underlag för att diskutera beteendeträds styrkor och svagheter i jämförelse med tidigare arbeten. Användningen av beteendeträd för att förena den mänskliga operatören med det autonoma systemet verkar lovande. Den föreslagna metoden att använda beteendeträd för att styra en svärm är tänkt att användas som ett gemensamt underlag för att möjliggöra att operatörer och ingenjörer kan ha en gemensam bild hur en svärm ska fungera.