Range finding are widely used in various industrial applications, such as machine vision, collision avoidance, and robotics. Presently most range finders either rely on active transmitters or sophisticated mechanical controllers and powerful processors to extract range information, which make the range finders costly, bulky, or slowly, and limit their applications. This dissertation is a detailed description of a real-time vision-based range sensing technique and its single-chip CMOS implementation. To the best of our knowledge, this system is the first single chip vision-based range finder that doesn't need any mechanical position adjustment, memory or digital processor. The entire signal processing on the chip is purely analog and occurs in parallel. The chip captures the image of an object and extracts the depth and range information from just a single picture. The on-chip, continuous-time, logarithmic photoreceptor circuits are used to couple spatial image signals into the range-extracting processing network. The photoreceptor pixels can adjust their operating regions, simultaneously achieving high sensitivity and wide dynamic range. The image sharpness processor and Winner-Take-All circuits are characterized and analyzed carefully for their temporal bandwidth and detection performance. The mathematical and optical models of the system are built and carefully verified. A prototype based on this technique has been fabricated and tested. The experimental results prove that the range finder can achieve acceptable range sensing precision with low cost and excellent speed performance in short-to-medium range coverage. Therefore, it is particularly useful for collision avoidance.

Many models exist in the aerospace industry that attempt to replicate the National Airspace System (NAS). The complexity of the NAS makes it a system that can be modeled in a variety of ways. While some NAS models are very detailed and take many factors into account, runtime of these simulations can be on the magnitude of hours (to simulate a single day). Other models forgo details in order to decrease the runtime of their simulation. Most models are capable of simulating a 24 hour period in the NAS. An analysis of an entire year would mean running the simulation for every day in the year, which would result in a long run time. The following thesis work presents a tool that is capable of giving the user a day that can be used in a simulation and will produce results similar to simulating the entire year. Taking in parameters chosen by the user, the tool outputs a single day, multiple days, or a composite day (based on percentages of days). Statistical methods were then used to compare each day to the overall year. On top of finding a single representative day, the ability to find a composite day was added. After implementing a brute force search technique to find the composite day, the long runtime was deemed inconvenient for the user. To solve this problem, a heuristic search method was created that would search the solution space in a short time and still output a composite day that represented the year. With a short runtime, the user would be able to run the program multiple times. Once the heuristic method was implemented, it was found that it performed well enough to make it an option for the user to choose. The final version of this tool was used to find a representative day and the result was used in comparison with output data from a NAS simulation model. Because the tool found the representative day based on historical data, it could be used to validate the effectiveness of the simulation model. The following thesis will go into detail about how this tool, the Representative Day Finder, was created.

This research presents an approach in assisting the control and operation of a rehabilitation robot manipulator to execute simple grasping tasks for persons with severe disabilities. It outlines the development of an end-effector sensory suite that includes the BarrettHand end-effector, laser range finder, and a low cost camera. The approach taken in this research differs greatly from the currently available rehabilitation robot arms in that it requires minimal user instruction, it is easy to operate and more effective for persons severely disabled. A thorough study of the currently available systems; Manus, Raptor and Kares II arm, is also presented. In order to test the end-effector sensory suite, experiments were performed to find the centroid of an object of interest to direct the robot end-effector towards it with minimal error. Analyses of centroid location data to ensure accurate results are also presented. The long term goal of this research is to significantly enhance the ability of severely disabled persons to perform activities of daily living using wheelchair mounted robot arms. The sensory suite developed through this project is expected to be integrated into a seven-degree of freedom wheelchair mounted robot arm currently under development at the Rehabilitation Robots Laboratory at the University of South Florida.

A direction-finding system with multiple direction finders (DFs) is a military intelligence system designed to detect the positions of transmitters of radio frequencies. This dissertation studies three decision problems associated with the direction-finding system. The first part of this dissertation is to prescribe DF deployment to maximize the effectiveness with which transmitter positions are estimated in an area of interest (AOI). Three methods are presented to prescribe DF deployment. The first method uses Stansfield’s probability density function to compute objective function coefficients numerically. The second and the third employ surrogate measures of effectiveness as objective functions. The second method, like the first, involves complete enumerations; the third formulates the problem as an integer program and solves it with an efficient network-based label-setting algorithm. Our results show that the third method, which involved use of a surrogate measure as an objective function and an exact label-setting algorithm, is most effective. The second part of this dissertation is to minimize the number of DFs to cover an AOI effectively, considering obstacles between DFs and transmitters. We formulate this problem as a partial set multicover problem in which at least -fraction of the likely transmitter positions must be covered, each by at least direction finders. We present greedy heuristics with random selection rules for the partial set multicover problem, estimating statistical bounds on unknown optimal values. Our results show that the greedy heuristic with column selection rule, which gives priority for selecting a column that advances more rows to k-coverage, performs best on the partial set multicover problems. Results also show that the heuristic with random row and column selection rules is the best of the heuristics with respect to statistical bounds. The third part of this dissertation deals with the problem of deploying direction finders with the goal of maximizing the effectiveness with which transmitter positions can be estimated in an AOI while hedging against enemy threats. We present four formulations, considering the probability that a direction finder deployed at a location will survive enemy threats over the planning horizon (i.e., not be rendered inoperative by an attack). We formulate the first two as network flow problems and present an efficient label-setting algorithm. The third and the fourth use the well-known Conditional Value at Risk (CVaR) risk measure to deal with the risk of being rendered inoperative by the enemy. Computational results show that risk-averse decision models tend to deploy some or all DFs in locations that are not close to the enemy to reduce risk. Results also show that a direction-finding system with 5 DFs provides improved survivability under enemy threats.

abstract: This experiment uses the Community of Knowledge framework to better understand how jurors interpret new information (Sloman & Rabb, 2016). Participants learned of an ostensibly new scientific finding that was claimed to either be well-understood or not understood by experts. Despite including no additional information, expert understanding led participants to believe that they personally understood the phenomenon, with expert understanding acting as a cue for trustworthiness and believability. This effect was particularly pronounced with low-quality sources. These results are discussed in the context of how information is used by jurors in court, and the implications of the “Community of Knowledge” effect being used by expert witnesses. / Dissertation/Thesis / Masters Thesis Psychology 2018

Context. A minority of criminals performs a majority of the crimes today. It is known that every criminal or group of offenders to some extent have a particular pattern (modus operandi) how crime is performed. Therefore, computers' computational power can be employed to discover crimes that have the same model and possibly are carried out by the same criminal. The goal of this thesis was to apply the existing Series Finder algorithm to a feature-rich dataset containing data about Swedish residential burglaries. Objectives. The following objectives were achieved to complete this thesis: Modifications performed on an existing Series Finder implementation to fit the Swedish police forces dataset and MatLab code converted to Python. Furthermore, experiment setup designed with appropriate metrics and statistical tests. Finally, modified Series Finder implementation's evaluation performed against both Spatial-Temporal and Random models. Methods. The experimental methodology was chosen in order to achieve the objectives. An initial experiment was performed to find right parameters to use for main experiments. Afterward, a proper investigation with dependent and independent variables was conducted. Results. After the metrics calculations and the statistical tests applications, the accurate picture revealed how each model performed. Series Finder showed better performance than a Random model. However, it had lower performance than the Spatial-Temporal model. The possible causes of one model performing better than another are discussed in analysis and discussion section. Conclusions. After completing objectives and answering research questions, it could be clearly seen how the Series Finder implementation performed against other models. Despite its low performance, Series Finder still showed potential, as presented in future work.

Monte Carlo Localization (MCL) is an algorithm that allows a robot to determine its location when provided a map of its surroundings. Particles, consisting of a location and an orientation, represent possible positions where the robot could be on the map. The probability of the robot being at each particle is calculated based on sensor input. Traditionally, MCL only utilizes the position of objects for localization. This thesis explores using wall and floor surface textures to help the algorithm determine locations more accurately. Wall textures are captured by using a laser range finder to detect patterns in the surface. Floor textures are determined by using an inertial measurement unit (IMU) to capture acceleration vectors which represent the roughness of the floor. Captured texture data is classified by an artificial neural network and used in probability calculations. The best variations of Texture MCL improved accuracy by 19.1\% and 25.1\% when all particles and the top fifty particles respectively were used to calculate the robot's estimated position. All implementations achieved comparable performance speeds when run in real-time on-board a robot.

This master’s thesis elaborates on the control of a wheeled mobile robot and its sensor system. The sensors provide information about distances of obstacles in its surroundings and orientation in area. The communication is realized by radiomodems, commands and data are sent by serial link between microncontroller and the PC.

An important post-9/11 objective has been to connect law enforcement agencies so they can share information that is routinely collected by police. This low-level information, gathered from sources such as traffic tickets, calls for service, incident reports and field contacts, is not widely shared but might account for as much as 97% of the data held in police records systems. U.S. policy and law assume that access to this information advances crime control and counterterrorism efforts. The scarcity of functioning systems has limited research opportunities to test this assumption or offer guidance to police leaders considering investments in information sharing. However, this study had access to FINDER, a Florida system that shares low-level data among 121 police agencies. The user-level value of FINDER was empirically examined using Goodhue's (1995) Task-Technology Fit framework. Objective system data from 1,352 users, user-reported "successes," and a survey of 402 active users helped define parameters of user-level success. Of the users surveyed, 68% reported arrests or case clearances, 71% reported improved performance, and 82% reported improved efficiency attributed to FINDER. Regression models identified system use, task-fit, and user characteristic measures that predicted changes in users' individual performance. A key finding was that FINDER affirmed the importance of sharing low-level police data, and successful outcomes were related to its ease of use and access to user-specified datasets. Also, users employed a variety of information-seeking techniques that were related to their task assignments. Improved understanding of user-defined success and system use techniques can inform the design and functionality of information sharing systems. Further, this study contributes to addressing the critical requirement for developing information sharing system metrics.

Multiple Sequenzalignments (MSAs) von homologen Proteinen sind nützliche Werkzeuge, um kompensatorische Mutationen zwischen nicht-konservierten Residuen zu charakterisieren. Die Identifizierung dieser Residuen in MSAs ist eine wichtige Aufgabe um die strukturellen Grundlagen und molekularen Mechanismen von Proteinfunktionen besser zu verstehen. Trotz der vielen Anzahl an Literatur über kompensatorische Mutationen sowie über die Sequenzkonservierungsanalyse für die Erkennung von wichtigen Residuen, haben vorherige Methoden meistens die biochemischen Eigenschaften von Aminosäuren nicht mit in Betracht gezogen, welche allerdings entscheidend für die Erkennung von kompensatorischen Mutationssignalen sein können. Jedoch werden kompensatorische Mutationssignale in MSAs oft durch das Rauschen verfälscht. Aus diesem Grund besteht ein weiteres Problem der Bioinformatik in der Trennung signifikanter Signale vom phylogenetischen Rauschen und beziehungslosen Paarsignalen. Das Ziel dieser Arbeit besteht darin Methoden zu entwickeln, welche biochemische Eigenschaften wie Ähnlichkeiten und Unähnlichkeiten von Aminosäuren in der Identifizierung von kompensatorischen Mutationen integriert und sich mit dem Rauschen auseinandersetzt. Deshalb entwickeln wir unterschiedliche Methoden basierend auf klassischer- und quantum Informationstheorie sowie multiple Testverfahren. Unsere erste Methode basiert auf der klassischen Informationstheorie. Diese Methode betrachtet hauptsächlich BLOSUM62-unähnliche Paare von Aminosäuren als ein Modell von kompensatorischen Mutationen und integriert sie in die Identifizierung von wichtigen Residuen. Um diese Methode zu ergänzen, entwickeln wir unsere zweite Methode unter Verwendung der Grundlagen von quantum Informationstheorie. Diese neue Methode unterscheidet sich von der ersten Methode durch gleichzeitige Modellierung ähnlicher und unähnlicher Signale in der kompensatorischen Mutationsanalyse. Des Weiteren, um signifikante Signale vom Rauschen zu trennen, entwickeln wir ein MSA-spezifisch statistisches Modell in Bezug auf multiple Testverfahren. Wir wenden unsere Methode für zwei menschliche Proteine an, nämlich epidermal growth factor receptor (EGFR) und glucokinase (GCK). Die Ergebnisse zeigen, dass das MSA-spezifisch statistische Modell die signifikanten Signale vom phylogenetischen Rauschen und von beziehungslosen Paarsignalen trennen kann. Nur unter Berücksichtigung BLOSUM62-unähnlicher Paare von Aminosäuren identifiziert die erste Methode erfolgreich die krankheits-assoziierten wichtigen Residuen der beiden Proteine. Im Gegensatz dazu, durch die gleichzeitige Modellierung ähnlicher und unähnlicher Signale von Aminosäurepaare ist die zweite Methode sensibler für die Identifizierung von katalytischen und allosterischen Residuen.