Monocular cameras are prominently used for estimating motion of Unmanned Aerial Vehicles. With growing interest in autonomous vehicle technology, the use of monocular cameras in ground vehicles is on the rise. This is especially favorable for localization in situations where Global Navigation Satellite System (GNSS) is unreliable, such as open-pit mining environments. However, most monocular camera based approaches suffer due to obscure scale information. Ground vehicles impose a greater difficulty due to high speeds and fast movements. This thesis aims to estimate the scale of monocular vision data by using an inertial sensor in addition to the camera. It is shown that the simultaneous estimation of pose and scale in autonomous ground vehicles is possible by the fusion of visual and inertial sensors in an Extended Kalman Filter (EKF) framework. However, the convergence of scale is sensitive to several factors including the initialization error. An accurate estimation of scale allows the accurate estimation of pose. This facilitates the localization of ground vehicles in the absence of GNSS, providing a reliable fall-back option. / Monokulära kameror används ofta vid rörelseestimering av obemannade flygande farkoster. Med det ökade intresset för autonoma fordon har även användningen av monokulära kameror i fordon ökat. Detta är fram för allt fördelaktigt i situationer där satellitnavigering (Global Navigation Satellite System (GNSS)) äropålitlig, exempelvis i dagbrott. De flesta system som använder sig av monokulära kameror har problem med att estimera skalan. Denna estimering blir ännu svårare på grund av ett fordons större hastigheter och snabbare rörelser. Syftet med detta exjobb är att försöka estimera skalan baserat på bild data från en monokulär kamera, genom att komplettera med data från tröghetssensorer. Det visas att simultan estimering av position och skala för ett fordon är möjligt genom fusion av bild- och tröghetsdata från sensorer med hjälp av ett utökat Kalmanfilter (EKF). Estimeringens konvergens beror på flera faktorer, inklusive initialiseringsfel. En noggrann estimering av skalan möjliggör också en noggrann estimering av positionen. Detta möjliggör lokalisering av fordon vid avsaknad av GNSS och erbjuder därmed en ökad redundans.

Autonomous driving systems are rapidly improving and may have the ability to change society in the coming decade. One important part of these systems is the interpretation of sensor information into trajectories of objects. In this master’s thesis, we study an energy minimisation method with radar and camera measurements as inputs. An energy is associated with the trajectories; this takes the measurements, the objects’ dynamics and more factors into consideration. The trajectories are chosen to minimise this energy, using a gradient descent method. The lower the energy, the better the trajectories are expected to match the real world. The processing is performed offline, as opposed to in real time. Offline tracking can be used in the evaluation of the sensors’ and the real time tracker’s performance. Offline processing allows for the use of more computer power. It also gives the possibility to use data that was collected after the considered point in time. A study of the parameters of the used energy minimisation method is presented, along with variations of the initial method. The results of the method is an improvement over the individual inputs, as well as over the real time processing used in the cars currently. In the parameter study it is shown which components of the energy function are improving the results. / Mycket resurser läggs på utveckling av självkörande bilsystem. Dessa kan komma att förändra samhället under det kommande decenniet. En viktig del av dessa system är behandling och tolkning av sensordata och skapande av banor för objekt i omgivningen. I detta examensarbete studeras en energiminimeringsmetod tillsammans med radar- och kameramätningar. En energi beräknas för banorna. Denna tar mätningarna, objektets dynamik och fler faktorer i beaktande. Banorna väljs för att minimera denna energi med hjälp av gradientmetoden. Ju lägre energi, desto bättre förväntas banorna att matcha verkligheten. Bearbetning sker offline i motsats till i realtid; offline-bearbetning kan användas då prestandan för sensorer och realtidsbehandlingen utvärderas. Detta möjliggör användning av mer datorkraft och ger möjlighet att använda data som samlats in efter den aktuella tidpunkten. En studie av de ingående parametrarna i den använda energiminimeringsmetoden presenteras, tillsammans med justeringar av den ursprungliga metoden. Metoden ger ett förbättrat resultat jämfört med de enskilda sensormätningarna, och även jämfört med den realtidsmetod som används i bilarna för närvarande. I parameterstudien visas vilka komponenter i energifunktionen som förbättrar metodens prestanda.

Recent advances in the telecommunications industry and the resulting applicationssuch as autonomous vehicles, vehicle surveillance and traffic safetyhas increased the demand for accurate and robust vehicle positioning systems.Existing Global Navigation Satellite System (GNSS) based positioning techniquesface significant performance loss in the tunnels and urban canyons.Recent researches have shown that radio-based positioning techniques are theoreticallypromising to make an accurate navigation system to fill the GNSSgaps. Fifth generation of mobile communication (5G) will utilize wide bandwidthstogether with beamforming enabled by antenna arrays to provide higherdata rates to mobile users. These features make 5G a favorable candidate forhigh accuracy positioning. On the other hand, sensor fusion is commonly employedto develop more robust and accurate navigation systems for vehicles. Inthis work, the range and angle measurements from 5G base stations are fusedwith the acceleration measurements by the means of the extended Kalman filterto generate position estimates for a moving car. The accuracy of this positioningsystem is studied with centimeter wave (cmWave) and millimeter wave(mmWave) 5G cellular networks which are set up by practical parameters. Towardsthat, the positioning system is tested in a simulation-based experimentwhere a car is moving on a highway and the 5G base stations are deployedalongside of it. Based on that, a detailed analysis of the Kalman filter’s rootmean squared error (RMSE) and the 5G’s different parameters and limitingfactors such as the line of sight (LOS) blockage is carried out. Our numericalresults show that vehicles connected to 5G can benefit from this system to enhancethe robustness and accuracy of their navigation system. / De senaste framstegen inom telekommunikationsindustrin och de resulterandeapplikationerna som autonoma fordon, fordonsövervakning och trafiksäkerhethar ökat efterfrågan på exakta fordonspositioneringssystem. ExisterandeGlobal Navigation Satellite System (GNSS) baserade positioneringsteknikerhar en betydande prestandaförlust i tunnlar och urbana kanjoner. Forskninghar visat att radiobaserade positioneringstekniker har mindre distributionskostnaderoch kan vara mer exakta än satellitbaserade navigationssystem.I den femte generation av mobilkommunikation (5G) används tekniker sommillimeterWave (mmWave) och multiple-input multiple-output (MIMO) därradio-terminaler består av stora matrisantenner och arbetar med stora bandbredder.Dessa funktioner gör 5G-system gynnsamma för positionering medhög noggrannhet. Å andra sidan har informationsfusion av Inertial NavigationSystems (INS) och andra positioneringstekniker vanligen använts för attutveckla mer robusta och exakta spårningssystem. I denna studie föreslår viett INS/5G-positioneringssystem för att spåra landfordon baserat på Kalmanfiltret. Vi adresserar systempositioneringsgränserna i termer av 5G nya radio(NR) subsystem och en detaljerad analys av beroendet av rotmedelfelteradkvadratfel (RMSE) för olika systemparametrar som utförs. Systemet testas iett enkelt simuleringsbaserat experiment som består av en rak motorväg medbasstationerna placerade bredvid det. Slutligen visar våra numeriska resultatatt det föreslagna systemet är i stånd att lokalisera ett UE-monterat fordon medsub-meter lägesfel även i närvaro av hård siktlinje blockering.

This thesis presents methods for robust precision landing of unmanned air vehicles (UAVs) on platforms at sea. Localization methods are proposed for UAV-to-boat state estimation for systems that employ real- time kinematic (RTK) global navigation satellite system (GNSS) and vision sensors. Solutions for GNSS-only are first presented, followed by the fusion of GNSS and vision. The important problem of sensor intrinsic calibration is solved with a novel offline batch estimation approach. Hardware results are presented for all methods. Our calibration of GNSS-to-camera is shown to estimate sensor offsets with millimeter level accuracy. Localization systems are combined with custom state machines that manage the landing attempt via a novel descent cone. This conical threshold enforces a safe and accurate landing. Our landing methods are demonstrated in real-world experiments and achieve consistent accurate landings with error below 10 cm. The fusion of camera and RTK is shown to produce a robust landing system with redundant localization sources.

No description available.

Myoelectric control schemes aim to utilize the surface electromyography (EMG) signals which are the electric potentials directly measured from skeletal muscles to control wearable robots such as exoskeletons and prostheses. The main challenge of myoelectric controls is to increase and preserve the signal quality by minimizing the effect of confounding factors such as muscle fatigue or electrode shift. Current research in myoelectric control schemes are developed to work in ideal laboratory conditions, but there is a persistent need to have these control schemes be more robust and work in real-world environments. Following the manifold hypothesis, complexity in the world can be broken down from a high-dimensional space to a lower-dimensional form or representation that can explain how the higher-dimensional real world operates. From this premise, the biological actions and their relevant multimodal signals can be compressed and optimally pertinent when performed in both laboratory and non-laboratory settings once the learned representation or manifold is discovered. This thesis outlines a method that incorporates the use of a contrastive variational autoencoder with an integrated classifier on multimodal sensor data to create a compressed latent space representation that can be used in future myoelectric control schemes.

Wireless sensor networks (WSNs) typically consist of a significant number of inexpensive sensor nodes, each of which is powered by a battery or another finite energy source that is difficult to replace because of the environment they are in or the cost of doing so. The applications of WSNs include military surveillance, disaster management, target tracking and monitoring environmental conditions. In order to increase the lifespan of WSNs, energy-efficient sensing and communication approaches for sensor nodes are essential. Recently, there has been an increase in interest in using unmanned aerial vehicles (UAVs) as portable data collectors for ground sensor nodes in WSN. Several approaches to solving effective communication between sensor nodes and the fusion center have been investigated in this thesis. Because processing, sensing range, transmission bandwidth, and energy consumption are always limited, it is beneficial not to use all the information provided at each sensor node in order to prolong its life span and reduce communication costs. In order to address this problem, first, efficient measurement quantization techniques are proposed using a single fusion center and multiple sensors. The dynamic bit distribution is done among all the sensors and within the measurement elements. The problem is then expanded to include multiple fusion centers, and a novel algorithm is proposed to associate sensors to fusion centers. The bandwidth distribution for targets which are being monitored by several sensors is addressed. Additionally, how to use the situation in which the sensors are in the coverage radius of multiple fusion centers in order to share the targets between them is discussed. Finally, performance bounded data collection algorithms are proposed where the necessary accuracy for each target is specified. In order to determine the minimum number of data collectors needed and their initial placement, an algorithm is proposed. When there are fewer fixed data collectors than there are regions to collect the data from, a coverage path planning method is developed. Since the optimal solution requires an enormous computational requirement and not realistic for real-time online implementation, approximate algorithms are proposed for multi-objective integer optimization problems. In order to assess each suggested algorithm's effectiveness, many simulated scenarios are used together with baselines and simple existing methods. / Thesis / Doctor of Philosophy (PhD)

Autonomous vehicles require intelligent systems to perceive and navigate unstructured envi- ronments. The scope of this project is to improve and develop algorithms and methods to support autonomy in the off-road problem space. This work explores computer vision architectures to support real-time object detection. Furthermore, this project explores multimodal deep fusion and sensor processing for off-road object detection. The networks are compared to and based off of the SqueezeSeg architecture. The MAVS simulator was utilized for data collection and semantic ground truth. The results indicate improvements from the SqueezeSeg performance metrics.

The need for higher-level reasoning capabilities beyond low-level sensor abilities has prompted researchers to use different types of sensor fusion techniques for better situational awareness in the intrusion detection environment. These techniques primarily vary in terms of their mission objectives. Some prioritize alerts for alert reduction, some cluster alerts to identify common attack patterns, and some correlate alerts to identify multi-staged attacks. Each of these tasks has its own merits. Unlike previous efforts in this area, this dissertation combines the primary tasks of sensor alert fusion, i.e., alert prioritization, alert clustering and alert correlation into a single framework such that individual results are used to quantify a confidence score as an overall assessment for global diagnosis of a system?s security health. Such a framework is especially useful in a multi-sensor environment where the sensors can collaborate with or complement each other to provide increased reliability, making it essential that the outputs of the sensors are fused in an effective manner in order to provide an improved understanding of the security status of the protected resources in the distributed environment. This dissertation uses a possibilistic approach in intelligent fusion of sensor alerts with Fuzzy Cognitive Modeling in order to accommodate the impreciseness and vagueness in knowledge-based reasoning. We show that our unified architecture for sensor fusion provides better insight into the security health of systems. A new multi-level alert clustering method is developed to accommodate inexact matching in alert features and is shown to provide relevance to more alerts than traditional exact clustering. Alert correlation with a new abstract incident modeling technique is shown to deal with scalability and uncertainty issues present in traditional alert correlation. New concepts of dynamic fusion are presented for overall situation assessment, which a) in case of misuse sensors, combines results of alert clustering and alert correlation, and b) in case of anomaly sensors, corroborates evidence from primary and secondary sensors for deriving the final conclusion on the systems? security health.

No description available.