Introduction: Medical errors are major hazards, and lapses in non-technical skills such as situational awareness contribute to most incidents. Risks are concentrated in acute care, and in crisis situations clinicians can apparently ignore vital information. Poor workplace ergonomics contributes to risk. Existing work into perceptual errors offers insights, but these phenomena have been little researched in medicine. This thesis considers medical non-technical skills and how they are taught, and explores vulnerability to inattentional and change blindness. Methods: Medical human factors and the psychology of perceptual error were reviewed, and a mixed-methods assessment of postgraduate medical curricula completed. Experiments assessed clinicians' interaction with clinical monitoring devices using eye-tracking, and studies were conducted exposing clinicians to various perceptual error stimuli using non-clinical and clinical videos, and simulation. A survey was also conducted to assess clinicians' insight into the phenomena of perceptual error. Results: Non-technical skills feature poorly in medical curricula, and equipment is poorly standardised in critical care areas. Unfamiliar devices slow response times and increase error rate. Clinical training confers no generalisable advantage in perceptual reliability. Even expert clinicians miss important events. Two out of every three life-support instructors for example missed a critical failure in the patient's oxygen supply when watching a recorded emergency simulation. The insight and understanding healthcare staff have of perceptual errors is poor, leading to significant overestimates of perceptual reliability that could have consequences for clinical practice. Conclusions: Perceptual errors represent a latent risk factor contributing to loss of situational awareness. High rates of perceptual error were observed in the video-based experiment. Although lower rates were observed in simulation, important events were still missed by participants that could have serious consequences. The incidence of perceptual error appears sensitive to the method used to test for it, and this has important implications for the design of future experiments testing for these phenomena. Mitigating perceptual error is likely to be challenging, but relatively simple adjustments to team practices in emergency situations may be fruitful.

[ES] Esta tesis doctoral propone el desarrollo de una arquitectura para sistema de detección de actividades criminales en vídeo aplicado a sistemas de mando y control para seguridad ciudadana. Este sistema está basado en la técnica de Deep Learning Faster R-CNN y tiene el novedoso enfoque de tratar las acciones criminales como los hurtos callejeros, en donde pueden ser identificados objetos como evidencia en una escena de vídeo. Esta tesis muestra el desarrollo de dicha aplicación, que demuestra ser efectiva, identificando la manera de reducir el costo computacional del análisis de vídeo cuadro a cuadro obteniendo rendimientos congruentes con las tasas de cuadros por segundo generados por cámaras de sistema de vídeo vigilancia ciudadana. También es objeto de estudio una posible implementación en el sistema de seguridad ciudadana de la Policía Nacional de Colombia. / [EN] This doctoral thesis proposes the development of a system to detect criminal activities in video applied to command and control systems for citizen security. This system is based on the Deep Learning technique called Faster R-CNN and has the novel approach of treating criminal actions like street thefts as objects that can be identified in a video scene. This thesis shows the development of this application and the way to reduce the computational cost of the video analysis frame by frame, obtaining performances congruent with the frame rate generated by citizen video surveillance system cameras. There is also a possible implementation in the citizen security system of the National Police of Colombia is being studied. / [CA] Esta tesi doctoral proposa el desenrotllament d'una arquitectura per a sistema de detecció d'activitats criminals en vídeo aplicat a sistemes de comandament i control per a seguretat ciutadana. Este sistema està basat en la tècnica de Deep Learning Faster R-CNN i té el nou enfocament de tractar les accions criminals com les afanades guies de carrers com a objectes que poden ser identificats en una escena de vídeo. Esta tesi mostra el desenrotllament de la dita aplicació, que demostra ser efectiva, identificant la manera de reduir el cost computacional de l'anàlisi de vídeo quadro a quadro obtenint rendiments congruents amb les taxes de cuados per segon generats per cambres de sistema de vídeo vigilància ciutadana. També s'estudia una possible implementació en el sistema de seguretat ciutadana de la Policia Nacional de Colòmbia. / Suárez Páez, JE. (2020). Arquitectura de detección de actividades criminales basada en análisis de vídeo en tiempo real [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/153162 / TESIS

<div> Space operations from the dawn of the Space Age have resulted in a large, and growing, resident space object population. However, the availability of sensor resources is limited, which presents a challenge to Space Situational Awareness applications. When direct communication with an object is not possible, whether that is due to a lack of access for active satellites or due to the object being characterized as debris, the only independent information source for learning about the resident space object population comes from measurements. Optical measurements are often a cost-effective method for obtaining information about resident space objects.<br></div><div> This work uses observability analysis to investigate the relationship between desired resident space object characteristics and the information resulting from ground-based optical measurements. Observability is a concept developed in modern control theory for evaluating whether the information contained within measurements is sufficient to describe the dynamical progression of a system over time. In this work, observability is applied to Space Situational Awareness applications to determine what object characteristic information can be recovered from ground-based optical measurements and under which conditions these determinations are possible. In addition, the constraints and limitations of applying observability to Space Situational Awareness applications are assessed and quantified.<br></div>

Complex Event Processing (CEP) systems aim at processing large flows of events to discover situations of interest (SOI). Primarily, CEP uses predefined pattern templates to detect occurrences of complex events in an event stream. Extracting complex event is achieved by employing techniques such as filtering and aggregation to detect complex patterns of many simple events. In general, CEP systems rely on domain experts to de fine complex pattern rules to recognize SOI. However, the task of fine tuning complex pattern rules in the event streaming environment face two main challenges: the issue of increased pattern complexity and the event streaming constraints where such rules must be acquired and processed in near real-time. Therefore, to fine-tune the CEP pattern to identify SOI, the following requirements must be met: First, a minimum number of rules must be used to re fine the CEP pattern to avoid increased pattern complexity, and second, domain knowledge must be incorporated in the refinement process to improve awareness about emerging situations. Furthermore, the event data must be processed upon arrival to cope with the continuous arrival of events in the stream and to respond in near real-time. In this dissertation, we present a Situation Refi nement Model (SRM) that considers these requirements. In particular, by developing a Single-Scan Frequent Item Mining algorithm to acquire the minimal number of CEP rules with the ability to adjust the level of re refinement to t the applied scenario. In addition, a cost-gain evaluation measure to determine the best tradeoff to identify a particular SOI is presented. / Graduate

In this project, an algorithm has been proposed to detect a satellite’s maneuver by comparingthe orbital elements observed from the two line element data and the orbital elements propagatedwith the help of Simplified perturbations models. A set of TLE data for an object orbiting Earthcontains a specific set of orbital elements. Simplified perturbation are utilized to propagate theorbital velocity and position vector of the same object. By comparing the results obtained fromboth the methods, the maneuvers of a satellite are detected. This project outlines the workingmethodology and the implementation of the algorithm developed to detect the maneuvers. Thefunctioning of the technique is assessed with reference to two case studies for which the maneuverhistory is available by following the approach employed by Kelecy et al. (2007). The same methodis implemented to detect the orbit controlling maneuvers as well as the fine control maneuvers. Theresults derived from the analysis indicate that the maneuvers which has the magnitude of even aslow as cm/s has been detected when the detection parameters are calibrated properly.

Network monitoring systems generate a high number of alerts reporting on anomalies and suspicious activity of IP addresses. From a huge number of alerts, only a small fraction is high priority and relevant from human evaluation. The rest is likely to be neglected. Assume that by analyzing large sums of these low priority alerts we can discover valuable information, namely, coordinated IP addresses and type of alerts likely to be correlated. This knowledge improves situational awareness in the field of network monitoring and reflects the requirement of security analysts. They need to have at their disposal proper tools for retrieving contextual information about events on the network, to make informed decisions. To validate the assumption new method is introduced to discover groups of coordinated IP addresses that exhibit temporal correlation in the arrival pattern of their events. The method is evaluated on real-world data from a sharing platform that accumulates 2.2 million alerts per day. The results show, that method indeed detected truly correlated groups of IP addresses.

This research introduces a satellite characterization method based on its light curve by utilizing and adapting the methodology of compressed sensing. Compressed sensing is a mathematical theory, which is established in signal compression and which has recently been applied to an image reconstruction by single-pixel camera observation. In this thesis, compressed sensing in the use of single-pixel camera observations is compared with a satellite characterization via non-resolved light curves. The assumptions, limitations, and significant differences in utilizing compressed sensing for satellite characterization are discussed in detail. Assuming a reference observation can be used to estimate the so-called sensing matrix, compressed sensing enables to approximately reconstruct resolved satellite images revealing details about the specific satellite that has been observed based solely on non-resolved light curves. This has been shown explicitly in simulations. This result implies the great potential of compressed sensing in characterizing space objects that are so far away that traditional resolved imaging is not possible.

<p>The motivation for this research stems from the promise of coupling multi-sensory systems and advanced data analytics to enhance holistic situational awareness and thus prevent fatal accidents in the construction industry. The construction industry is one of the most dangerous industries in the U.S. and worldwide. Occupational Safety and Health Administration (OSHA) reports that the construction sector employs only 5% of the U.S. workforce, but accounts for 21.1% (1,008 deaths) of the total worker fatalities in 2018. The struck-by accident is one of the leading causes and it alone led to 804 fatalities between 2011 and 2015. A critical contributing factor to struck-by accidents is the lack of holistic situational awareness, attributed to the complex and dynamic nature of the construction environment. In the context of construction site safety, situational awareness consists of three progressive levels: perception – to perceive the status of construction entities on the jobsites, comprehension – to understand the ongoing construction activities and interactions among entities, and projection – to predict the future status of entities on the dynamic jobsites. In this dissertation, holistic situational awareness refers to the achievement at all three levels. It is critical because with the absence of holistic situational awareness, construction workers may not be able to correctly recognize the potential hazards and predict the severe consequences, either of which will pose workers in great danger and may result in construction accidents. While existing studies have been successful, at least partially, in improving the perception of real-time states on construction sites such as locations and movements of jobsite entities, they overlook the capability of understanding the jobsite context and predicting entity behavior (i.e., movement) to develop the holistic situational awareness. This presents a missed opportunity to eliminate construction accidents and save hundreds of lives every year. Therefore, there is a critical need for developing holistic situational awareness of the complex and dynamic construction sites by accurately perceiving states of individual entities, understanding the jobsite contexts, and predicting entity movements.<br></p><p>The overarching goal of this research is to minimize the risk of struck-by accidents on construction jobsite by enhancing the holistic situational awareness of the unstructured and dynamic construction environment through a novel data-driven approach. Towards that end, three fundamental knowledge gaps/challenges have been identified and each of them is addressed in a specific objective in this research.<br></p> <p>The first knowledge gap is the lack of methods in fusing heterogeneous data from multimodal sensors to accurately perceive the dynamic states of construction entities. The congested and dynamic nature of construction sites has posed great challenges such as signal interference and line of sight occlusion to a single mode of sensor that is bounded by its own limitation in perceiving the site dynamics. The research hypothesis is that combining data of multimodal sensors that serve as mutual complementation achieves improved accuracy in perceiving dynamic states of construction entities. This research proposes a hybrid framework that leverages vision-based localization and radio-based identification for robust 3D tracking of multiple construction workers. It treats vision-based tracking as the main source to obtain object trajectory and radio-based tracking as a supplementary source for reliable identity information. It was found that fusing visual and radio data increases the overall accuracy from 88% and 87% to 95% and 90% in two experiments respectively for 3D tracking of multiple construction workers, and is more robust with the capability to recover the same entity ID after fragmentation compared to using vision-based approach alone.<br></p> <p>The second knowledge gap is the missing link between entity interaction patterns and diverse activities on the jobsite. With multiple construction workers and equipment co-exist and interact on the jobsite to conduct various activities, it is extremely difficult to automatically recognize ongoing activities only considering the spatial relationship between entities using pre-defined rules, as what has been done in most existing studies. The research hypothesis is that incorporating additional features such as attentional cues better represents entity interactions and advanced deep learning techniques automates the learning of the complex interaction patterns underlying diverse activities. This research proposes a two-step long short-term memory (LSTM) approach to integrate the positional and attentional cues to identify working groups and recognize corresponding group activities. A series of positional and attentional cues are modeled to represent the interactions among entities, and the LSTM network is designed to (1) classify whether two entities belong to the same group, and (2) recognize the activities they are involved in. It was found that by leveraging both positional and attentional cues, the accuracy increases from 85% to 95% compared with cases using positional cues alone. Moreover, dividing the group activity recognition task into a two-step cascading process improves the precision and recall rates of specific activities by about 3%-12% compared to simply conducting a one-step activity recognition.<br></p> <p>The third knowledge gap is the non-determining role of jobsite context on entity movements. Worker behavior on a construction site is goal-based and purposeful, motivated and influenced by the jobsite context including their involved activities and the status of other entities. Construction workers constantly adjust their movements in the unstructured and dynamic workspace, making it challenging to reliably predict worker trajectory only considering their previous movement patterns. The research hypothesis is that combining the movement patterns of the target entity with the jobsite context more accurately predicts the trajectory of the entity. This research proposes a context-augmented LSTM method, which incorporates both individual movement and workplace contextual information, for better trajectory prediction. Contextual information regarding movements of neighboring entities, working group information, and potential destination information is concatenated with movements of the target entity and fed into an LSTM network with an encoder-decoder architecture to predict trajectory over multiple time steps. It was found that integrating contextual information with target movement information can result in a smaller final displacement error compared to that obtained only considering the previous movement, especially when the length of prediction is longer than the length of observation. Insights are also provided on the selection of appropriate methods.<br></p><p>The results and findings of this dissertation will augment the holistic situational awareness of site entities in an automatic way and enable them to have a better understanding of the ongoing jobsite context and a more accurate prediction of future states, which in turn allows the proactive detection of any potential collisions.<br></p>

The resident space object population in the near-Earth vicinity has steadily increased since the dawn of the space age. This population is expected to increase drastically in the near future as the realization of proposed mega-constellations is already underway. The resultant congestion in near-Earth space necessitates the availability of more complete and more accurate satellite tracking information to ensure the continued sustainable use of this environment. This work sets out to create an operational system for the delivery of accurate satellite tracking information by means of optical observation. The state estimates resulting from observation series conducted on a GPS satellite and a geostationary satellite are presented and compared to existing catalog information. The satellite state estimate produced by the system is shown to outperform existing two-line element results. Additionally, the statistical information provided by the processing pipeline is evaluated and found to be representative of the best information available for the satellites true state.

Satellites commonly use onboard digital cameras, called star trackers. A star tracker determines the satellite's attitude, i.e. its orientation in space, by comparing star positions with databases of star patterns. In this thesis, I investigate the possibility of extending the functionality of star trackers to also detect the presence of resident space objects (RSO) orbiting the earth. RSO consist of both active satellites and orbital debris, such as inactive satellites, spent rocket stages and particles of different sizes. I implement and compare nine detection algorithms based on image analysis. The input is two hundred synthetic images, consisting of a portion of the night sky with added random Gaussian and banding noise. RSO, visible as faint lines in random positions, are added to half of the images. The algorithms are evaluated with respect to sensitivity (the true positive rate) and specificity (the true negative rate). Also, a difficulty metric encompassing execution times and computational complexity is used. The Laplacian of Gaussian algorithm outperforms the rest, with a sensitivity of 0.99, a specificity of 1 and a low difficulty. It is further tested to determine how its performance changes when varying parameters such as line length and noise strength. For high sensitivity, there is a lower limit in how faint the line can appear. Finally, I show that it is possible to use the extracted information to roughly estimate the orbit of the RSO. This can be accomplished using the Gaussian angles-only method. Three angular measurements of the RSO positions are needed, in addition to the times and the positions of the observer satellite. A computer architecture capable of image processing is needed for an onboard implementation of the method.