Digital Commons Network^™

Adaptable And Trustworthy Machine Learning For Human Activity Recognition From Bioelectric Signals, Morgan S. Stuart

Theses and Dissertations

Enabling machines to learn measures of human activity from bioelectric signals has many applications in human-machine interaction and healthcare. However, labeled activity recognition datasets are costly to collect and highly varied, which challenges machine learning techniques that rely on large datasets. Furthermore, activity recognition in practice needs to account for user trust - models are motivated to enable interpretability, usability, and information privacy. The objective of this dissertation is to improve adaptability and trustworthiness of machine learning models for human activity recognition from bioelectric signals. We improve adaptability by developing pretraining techniques that initialize models for later specialization to unseen …

Framing Automation Trust: How Initial Information About Automated Driving Systems Influences Swift Trust In Automation And Trust Repair For Human Automation Collaboration, Scott Anthony Mishler

Psychology Theses & Dissertations

The study examines how trust in automation is influenced by initial framing of information before interaction and how later active calibration methods can further influence trust repair or dampening after an automation error in a three-experiment study. As more human drivers begin to use automated driving systems (ADSs) for the first time, their initial understanding of the system can influence their trust leading to a miscalibration of trust. Prior studies have investigated how trust develops through interactions with an automated system, but few have looked at integrating swift trust and framing to calibrate trust before interaction and investigate further active …

Trust-Based Runtime Verification Of Autonomous Robotic Systems, Maziar Fooladi Mahani

All Dissertations

 Trust plays a crucial role in enabling effective collaboration and decision-making within human-multi-robot teams. In this context, runtime verification techniques and trust inference models have emerged as valuable tools for assessing and quantifying the trustworthiness of individual robots. This dissertation presents a study on trust-based runtime verification and introduces a Bayesian trust inference model for human-multi-robot teams. Firstly, we discuss the concept of runtime verification, which involves monitoring and analyzing the behavior of robots during their operation. We highlight the importance of trust as a key factor in determining the reliability and credibility of robot actions. By integrating trust metrics …

Defining Safe Training Datasets For Machine Learning Models Using Ontologies, Lynn C. Vonder Haar

Doctoral Dissertations and Master's Theses

Machine Learning (ML) models have been gaining popularity in recent years in a wide variety of domains, including safety-critical domains. While ML models have shown high accuracy in their predictions, they are still considered black boxes, meaning that developers and users do not know how the models make their decisions. While this is simply a nuisance in some domains, in safetycritical domains, this makes ML models difficult to trust. To fully utilize ML models in safetycritical domains, there needs to be a method to improve trust in their safety and accuracy without human experts checking each decision. This research proposes …

A Design Thinking Framework For Human-Centric Explainable Artificial Intelligence In Time-Critical Systems, Paul Benjamin Stone

Browse all Theses and Dissertations

Artificial Intelligence (AI) has seen a surge in popularity as increased computing power has made it more viable and useful. The increasing complexity of AI, however, leads to can lead to difficulty in understanding or interpreting the results of AI procedures, which can then lead to incorrect predictions, classifications, or analysis of outcomes. The result of these problems can be over-reliance on AI, under-reliance on AI, or simply confusion as to what the results mean. Additionally, the complexity of AI models can obscure the algorithmic, data and design biases to which all models are subject, which may exacerbate negative outcomes, …

Recognizing Potential Cyberspace Warriors Through The Use Of Suspicion Propensity Index, Meghan G. Strang

Theses and Dissertations

The Air Force currently suffers from excessively high attrition rates in the cyberspace career. The field is currently aiming to develop an entrance examination for the career, striving to improve personnel selection and decrease these high attrition rates. The attribute of suspicion is a key focus of the preliminary examination development, theorizing that it is a potential indicator present among competent cyber operators. This research makes use of the Suspicion Propensity Index (SPI), a reliable measure of one’s tendency to be suspicious, along with the highly cited Mayer’s trust questionnaire in order to compare the attributes that appear in successful …

Eye-Tracking To Evaluate Trust In Human-Atr Interaction, Samuel Francis Adelman

Browse all Theses and Dissertations

Human collaboration with targeting aids have allowed analysts to achieve a greater level of coordination and productivity in a variety of fields. This project investigates the impact that an Assisted Target Recognition (ATR) algorithm’s false alarm rate and the task Target of Interest (TOI) level has on user-system trust and use in a targeting decision task. Previous studies suggest that an increased number of false alarms in an ATR task negatively impacts analyst trust in the system. This study will further contribute to this research, aiming to provide a better framework for appropriate tolerance levels within ATR algorithms, utilizing pre-truthed …

Trust Based Time-Varying Network Topology For Distributed Co-Operative Control Of Multi-Class Multi-Agent Systems, Ankur Vipulkumar Dalal

Mechanical and Aerospace Engineering Theses

With increased levels of autonomy in most of the engineering fields and booms in areas such as swarms, platoons and Internet of Things (IoT), communication and information flow has become a highly researched field. With advancements in autonomous vehicles (AVs) and drones in armed warfare, more and more focus is being laid on intercommunication between these vehicles and its surroundings as well as intra-communication among the fleets/swarms itself. It is easier to deal with individual agents whereas it is quite challenging to deal with multi-agent systems especially with highly dynamic agents. In this thesis, we propose a general protocol for …

Towards Tools For Achieving Third-Party Ip Assurance, Sean Talbot Jensen

Theses and Dissertations

Intellectual Property (IP) is used to speed up the design process and save money. The use of IP and complex CAD tools reduce visibility into the design and what is actually happening during synthesis and implementation. All of the complexity makes it easier for an attacker to insert malicious logic or tamper with the design in ways that are difficult to detect. Not very much work has been done towards the creation of tools to facilitate the safe use of 3rd-party IP. This work presents Physical and Functional Assurance, two approaches that aim to accomplish this task through physically and …

The Effect Of Task Load, Automation Reliability, And Environment Complexity On Uav Supervisory Control Performance, Sarah M. Sherwood

Doctoral Dissertations and Master's Theses

Over the last decade, military unmanned aerial vehicles (UAVs) have experienced exponential growth and now comprise over 40% of military aircraft. However, since most military UAVs require multiple operators (usually an air vehicle operator, payload operator, and mission commander), the proliferation of UAVs has created a manpower burden within the U.S. military. Fortunately, simultaneous advances in UAV automation have enabled a switch from direct control to supervisory control; future UAV operators will no longer directly control a single UAV subsystem but, rather, will control multiple advanced, highly autonomous UAVs. However, research is needed to better understand operator performance in a …

Communication-Cognizant Control Of Microgrids, Fnu Shankar Abhinav

Electrical Engineering Dissertations

The increased integration of power electronic devices, advanced software-based controllers, and communication networks has transformed the microgrid from a simple physical system into a cyber-physical system. In this work, advanced distributed control algorithms and communication network are leveraged to provide a robust control structure for AC and DC microgrids. Introduction of distributed control and communication networks in microgrids presents its own unique challenges, which has not been considered in existing literature. A noise-resilient distributed controller is proposed for synchronization of frequency and voltage in AC microgrids. Synchronization in AC microgrids is cast as an optimization problem, and solved using Alternating …

Diversity Team Building: Impact On Virtual Team Performance, Nina C. Magpili-Smith

Engineering Management & Systems Engineering Theses & Dissertations

Although organizations have addressed diversity issues at the organizational-level with resulting positive employee outcomes, lack of scholarly attention to team-level interventions remain. Team-level interventions would benefit organizations more directly as they address issues directly related to task accomplishment. Since diversity may lead to negative performance results for teams, a team building intervention based on the latest empirical research was developed and tested to address the potential performance losses associated to diversity in decision-making teams. The team building intervention provides six crucial elements, namely (1) direct experience of how deep-level team diversity affect team dynamics, (2) diversity education, (3) cultivation of …

Abi And Beyond: Exploration Of The Precursors To Trust In The Human-Automation Domain, Christopher Stephen Calhoun

Browse all Theses and Dissertations

This study investigated the extent to which the well-known precursors of interpersonal trust (ability, benevolence, integrity, or ABI) could be exploited, redefined, or added to when considering and developing models of trust between humans and technology. The ABI model explains only about half of the variation in interpersonal trust (Colquitt, Scott, & LePine, 2007), so two additional precursors to trust from the interpersonal and automation trust domains - transparency and humanness - were identified and studied. The experimental task involved users interacting with an automated aid (image processing and recommender system) through a simulated unmanned ground vehicle (UGV) interface to …

An Empirical Investigation Of The Willingness Of Us Intelligence Community Analysts To Contribute Knowledge To A Knowledge Management System (Kms) In A Highly Classified And Sensitive Environment, Robert Hambly

CCE Theses and Dissertations

Since September 11, 2001, the United States Government (USG) has possessed unparalleled capability in terms of dedicated intelligence and information collection assets supporting the analysts of the Intelligence Community (IC). The USG IC has sponsored, developed, and borne witness to extraordinary advances in technology, techniques, and procedures focused on knowledge harvesting, knowledge sharing, and collaboration. Knowledge, within successful (effective & productive) organizations, exists as a commodity; a commodity that can be created, captured, imparted, shared, and leveraged. The research problem that this study addressed is the challenge of maintaining strong organizational effectiveness and productivity through the use of an information …

Co-Design Of Dynamic Real-Time Scheduling And Cooperative Control For Human-Agent Collaboration Systems Based On Mutual Trust, Xiaotian Wang

All Theses

Mutual trust is a key factor in human-human collaboration. Inspired by this social interaction, we analyze human-agent mutual trust in the collaboration of one human and (semi)autonomous multi-agent systems. In the thesis, we derive time-series human-agent mutual trust models based on results from human factors engineering. To avoid both over- trust and under-trust, we set up dynamic timing models for the multi-agent scheduling problem and develop necessary and sufficient conditions to test the schedulability of the human multi-agent collaborative task. Furthermore, we extend the collaboration between one human and multiple agents into the collaboration between multi-human network and swarm-based agents …

Analysis And Synthesis Of Effective Human-Robot Interaction At Varying Levels In Control Hierarchy, David Spencer

All Theses

Robot controller design is usually hierarchical with both high-level task and motion planning and low-level control law design. In the presented works, we investigate methods for low-level and high-level control designs to guarantee joint performance of human-robot interaction (HRI). In the first work, a low-level method using the switched linear quadratic regulator (SLQR), an optimal control policy based on a quadratic cost function, is used. By incorporating measures of robot performance and human workload, it can be determined when to utilize the human operator in a method that improves overall task performance while reducing operator workload. This method is demonstrated …

Influence Of Task-Role Mental Models On Human Interpretation Of Robot Motion Behavior, Scott Ososky

Electronic Theses and Dissertations

The transition in robotics from tools to teammates has begun. However, the benefit autonomous robots provide will be diminished if human teammates misinterpret robot behaviors. Applying mental model theory as the organizing framework for human understanding of robots, the current empirical study examined the influence of task-role mental models of robots on the interpretation of robot motion behaviors, and the resulting impact on subjective ratings of robots. Observers (N = 120) were exposed to robot behaviors that were either congruent or incongruent with their task-role mental model, by experimental manipulation of preparatory robot task-role information to influence mental models (i.e., …

A Multi-Parameter Functional Side Channel Analysis Method For Hardware Trojan Detection In Untrusted Fpga Bitstreams, Christopher William Bell

USF Tampa Graduate Theses and Dissertations

Hardware Trojan Horses (HTHs or Trojans) are malicious design modifications intended to cause the design to function incorrectly. Globalization of the IC development industry has created new opportunities for rogue agents to compromise a design in such a way. Offshore foundries cannot always be trusted, and the use of trusted foundries is not always practical or economical. There is a pressing need for a method to reliably detect these Trojans, to prevent compromised designs from being put into production.

This thesis proposes a multi-parameter analysis method that is capable of reliably detecting function-altering and performance-degrading Trojans in FPGA bitstreams. It …

Visualization Methods And User Interface Design Guidelines For Rapid Decision Making In Complex Multi-Task Time-Critical Environments, Sriram Mahadevan

Browse all Theses and Dissertations

Real-world scenarios are complex dynamic systems that are often overloaded with information. Effective performance of these dynamic systems depends on the objects in such systems and the relationship among them. The control of many of these systems is semi-automated. Human operators constantly monitor and control these systems, assess the situation and often make decisions under time pressure. However, this supervisory control paradigm in a dual-task environment can be a very challenging task. Existing interface design methodologies and techniques have not delved deeply enough into defining information displays for complex, dynamic, time-critical, dual-task environments with capabilities for rapid task change awareness …

The Effect Of External Safeguards On Human-Information System Trust In An Information Warfare Environment, Gregory S. Fields

Theses and Dissertations

This research looks at how human trust in an information system is influenced by external safeguards in an Information Warfare (1W) domain. Information systems are relied upon in command and control environments to provide fast and reliable information to the decision-makers. The degree of reliance placed in these systems by the decision-makers suggests a significant level of trust. Understanding this trust relationship and what effects it is the focus of this study. A model is proposed that predicts behavior associated with human trust in information systems. It is hypothesized that a decision-maker's belief in the effectiveness of external safeguards will …