Many domains, such as emergency assistance, agriculture, construction, and planetary exploration, will increasingly require effective coordination of teams of robots and humans to accomplish a collection of spatially distributed heterogeneous tasks. Such coordination problems range from those that require loosely coordinated teams in which agents independently perform their assigned tasks, to those that require tightly coordinated teams where all actions of the team members need to be tightly synchronized. The scenarios of interest to this thesis lie between these two extremes, where some tasks are independent and others are related by constraints such as precedence, simultaneity, or proximity. These constraints may be a result of different factors including the complementary capabilities of different types of agents which require them to cooperate to achieve certain goals. The manner in which the constraints are satisfied influences the overall utility of the team. This thesis explores the problem of task allocation, scheduling, and routing for heterogeneous teams with such cross-schedule dependencies. We first describe and position this coordination problem in the larger space of multi-robot task allocation problems and propose an enhanced taxonomy for this space of problems. Recognizing that solution quality is important in many domains, we then present a mathematical programming approach to computing a bounded-optimal solution to the task allocation, scheduling and routing problem with cross-schedule dependencies. Specifically, we present a branch-and-price algorithm operating on a set-partitioning formulation of the problem, with side constraints. This bounded optimal “anytime” algorithm computes progressively better solutions and bounds, until it eventually terminates with the optimal solution. By examining the behavior of this algorithm, we gain insight into the impact on problem difficulty of various problem features, particularly different types of cross-schedule dependencies. Lastly, the thesis presents a flexible execution strategy for the resulting team plans with cross-schedule dependencies, and results demonstrating the approach on a team of indoor robots

In the medical field images obtained from high definition cameras and other medical imaging systems are an integral part of medical diagnosis. The analysis of these images are usually performed by the physicians who sometimes need to spend long hours reviewing the images before they are able to come up with a diagnosis and then decide on the course of action. In this dissertation we present a framework for a computer-aided analysis of medical imagery via the use of an expert system. While this problem has been discussed before, we will consider a system based on mobile devices. Since the release of the iPhone on April 2003, the popularity of mobile devices has increased rapidly and our lives have become more reliant on them. This popularity and the ease of development of mobile applications has now made it possible to perform on these devices many of the image analyses that previously required a personal computer. All of this has opened the door to a whole new set of possibilities and freed the physicians from their reliance on their desktop machines. The approach proposed in this dissertation aims to capitalize on these new found opportunities by providing a framework for analysis of medical images that physicians can utilize from their mobile devices thus remove their reliance on desktop computers. We also provide an expert system to aid in the analysis and advice on the selection of medical procedure. Finally, we also allow for other mobile applications to be developed by providing a generic mobile application development framework that allows for access of other applications into the mobile domain. In this dissertation we outline our work leading towards development of the proposed methodology and the remaining work needed to find a solution to the problem. In order to make this difficult problem tractable, we divide the problem into three parts: the development user interface modeling language and tooling, the creation of a game development modeling language and tooling, and the development of a generic mobile application framework. In order to make this problem more manageable, we will narrow down the initial scope to the hair transplant, and glaucoma domains.

The self-organized nest construction behaviors of European paper wasps (Polistes dominula) show potential for adoption in artificial intelligence and robotic systems where centralized control proves challenging. However, P. dominula nest construction mechanisms are not fully understood. The goal of this research was to investigate how P. dominula nest structures stimulate worker actions. Simulation utilities were constructed in C++, C#, and Python. Two models from previous work, a three-dimensional model with weighted actions and a two-dimensional model with simple rule-based actions, were combined in a three-dimensional model with simple rules. Nest construction was simulated with a random selection rule, an age-based rule, a height requirement rule, and a height difference rule. Real and idealized nest data were used to evaluate simulated nests. Structures generated with age- and height-based rules showed more correlation with real and idealized nest structures than randomly-generated structures.

This study investigates the portrayal and perception of artificial intelligence (AI) in I, Robot (2004) and Chappie (2015), providing one of the first accounts of the causality between attitudes and expectations in the representation and reception of films about AI. The findings suggest that the level of optimism of a film is likely to be linked to its socio-cultural context. The humanoid representation of each robotic protagonist prevented each film from skewing too far towards the extremes of technological optimism or pessimism. This affected respondents’ attitudes immediately after viewership, but this affect was short-lived. Additionally, while portrayals of the future somewhat aligned to contemporary developments regarding weak AI, they were overly optimistic or pessimistic about the future of strong AI. This had little impact on respondents’ fears and expectations, as respondents used the films as visual aids to mentally depict abstract concepts relating to AI that were arrived at elsewhere. / Communication Science / M.A. (Communication Science)

Supervised machine learning models are increasingly being used for medical diagnosis. The diagnostic problem is formulated as a binary classification task in which trained classifiers make predictions based on a set of input features. In diagnosis, these features are typically procedures or tests with associated costs. The cost of applying a trained classifier for diagnosis may be estimated as the total cost of obtaining values for the features that serve as inputs for the classifier. Obtaining classifiers based on a low cost set of input features with acceptable classification accuracy is of interest to practitioners and researchers. What makes this problem even more challenging is that costs associated with features vary with patients and service providers and change over time. This dissertation aims to address this problem by proposing a method for obtaining low cost classifiers that meet specified accuracy requirements under dynamically changing costs. Given a set of relevant input features and accuracy requirements, the goal is to identify all qualifying classifiers based on subsets of the feature set. Then, for any arbitrary costs associated with the features, the cost of the classifiers may be computed and candidate classifiers selected based on cost-accuracy tradeoff. Since the number of relevant input features k tends to be large for typical diagnosis problems, training and testing classifiers based on all 2^k-1 possible non-empty subsets of features is computationally prohibitive. Under the reasonable assumption that the accuracy of a classifier is no lower than that of any classifier based on a subset of its input features, this dissertation aims to develop an efficient method to identify all qualifying classifiers. This study used two types of classifiers – artificial neural networks and classification trees – that have proved promising for numerous problems as documented in the literature. The approach was to measure the accuracy obtained with the classifiers when all features were used. Then, reduced thresholds of accuracy were arbitrarily established which were satisfied with subsets of the complete feature set. Threshold values for three measures –true positive rates, true negative rates, and overall classification accuracy were considered for the classifiers. Two cost functions were used for the features; one used unit costs and the other random costs. Additional manipulation of costs was also performed. The order in which features were removed was found to have a material impact on the effort required (removing the most important features first was most efficient, removing the least important features first was least efficient). The accuracy and cost measures were combined to produce a Pareto-Optimal Frontier. There were consistently few elements on this Frontier. At most 15 subsets were on the Frontier even when there were hundreds of thousands of acceptable feature sets. Most of the computational time is taken for training and testing the models. Given costs, models in the Pareto-Optimal Frontier can be efficiently identified and the models may be presented to decision makers. Both the Neural Networks and the Decision Trees performed in a comparable fashion suggesting that any classifier could be employed.

No description available.

Comic books are a unique and increasingly popular form of entertainment combining visual and textual elements of communication. This work pertains to making comic books more accessible. Specifically, this paper explains how we detect elements such as speech bubbles present in Japanese comic book panels. Some applications of the work presented in this paper are automatic detection of text and its transformation into audio or into other languages. Automatic detection of elements can also allow reasoning and analysis at a deeper semantic level than what’s possible today. Our approach uses an expert system and a machine learning system. The expert system process information from images and inspires feature sets which help train the machine learning system. The expert system detects speech bubbles based on heuristics. The machine learning system uses machine learning algorithms. Specifically, Naive Bayes, Maximum Entropy, and support vector machine are used to detect speech bubbles. The algorithms are trained in a fully-supervised way and a semi-supervised way. Both the expert system and the machine learning system achieved high accuracy. We are able to train the machine learning algorithms to detect speech bubbles just as accurately as the expert system. We also applied the same approach to eye detection of characters in the panels, and are able to detect majority of the eyes but with low precision. However, we are able to improve the performance of our eye detection system significantly by combining the SVM and either the Naive Bayes or the AdaBoost classifiers.

A recent trend among sports fans along both sides of the letterman jacket is that of Daily Fantasy Sports (DFS). The DFS industry has been under legal scrutiny recently, due to the view that daily sports data is too random to make its prediction skillful. Therefore, a common view is that it constitutes online gambling. This thesis proves that DFS, as it pertains to Baseball, is significantly more predictable than random chance, and thus does not constitute gambling. We propose a system which generates daily lists of lineups for Fanduel Daily Fantasy Baseball contests. The system consists of two components: one for predicting player scores for every player on a given day, and one for generating lists of the best combinations of players (lineups) using the predicted player scores. The player score prediction component makes use of deep neural network models, including a Long Short-Term Memory recurrent neural network, to model daily player performance over the 2016 and 2017 MLB seasons. Our results indicate this to be a useful prediction tool, even when not paired with the lineup generation component of our system. We build off of previous work to develop two models for lineup generation, one completely novel, dependent on a set of player predictions. Our evaluations show that these lineup generation models paired with player predictions are significantly better than random, and analysis shows insights into key aspects of the lineup generation process.

With the rising popularity of small satellites, such as CubeSats, many smaller institutions previously incapable of developing and deploying a spacecraft have starting to do so. Institutions with a history of space flight, such as NASA JPL, have begun to put projects on CubeSats that would normally fly on much larger satellites. As a result, the institutions with space flight heritage have begun to port spacecraft software that was previously designed for much larger and more complex satellites to the CubeSat platform. Unfortunately for universities, who are the majority of all institutions devel- oping CubeSats, these ported systems are too large and complex to be a practical control solution. Student teams have a high turnover rate due to graduation and when a student becomes an expert on the control system, they graduate; most students get a maximum of two or three years of experience before graduating. This thesis proposes the Generic Decision Making Framework for Autonomous Systems (GDMFAS) as an accessible, easily extensible, component-based executive system architecture. The architecture is designed for Linux distributions, including the custom Linux distribution used by PolySat, and is implemented using C++. The proposed framework provides much of the same functionality as systems designed for larger satellites in a smaller, more straightforward pack- age, which includes both scheduling and executive components. This thesis also provides validation for the prototype implementation and evaluates the system according to six metrics. The metric analysis for this work is then compared with the metric analyses of previous works.

The increase in popularity for wearable technologies (see: Apple Watch and Microsoft Band) has opened the door for an Internet of Things solution to healthcare. One of the most prevalent healthcare problems today is the poor survival rate of out-of hospital sudden cardiac arrests (9.5% on 360,000 cases in the USA in 2013). It has been proven that heart rate derived features can give an early indicator of sudden cardiac arrest, and that providing an early warning has the potential to save many lives. Many of these new wearable devices are capable of providing this warning through their heart rate sensors. This thesis paper introduces a prospective dataset of physical activity heart rates collected via Microsoft Band. This dataset is indicative of the heart rates that would be observed in the proposed Internet of Things solution. This dataset is combined with public heart rate datasets to provide a dataset larger than many of the ones used in related works and more indicative of out-of-hospital heart rates. This paper introduces the use of LogitBoost as a classifier for sudden cardiac arrest prediction. Using this technique, a five minute warning of sudden cardiac arrest is provided with 96.36% accuracy and F-score of 0.9375. These results are better than existing solutions that only include in-hospital data.