Engineering Commons^™

Understanding The Role Of Interactivity And Explanation In Adaptive Experiences, Lijie Guo

All Dissertations

Adaptive experiences have been an active area of research in the past few decades, accompanied by advances in technology such as machine learning and artificial intelligence. Whether the currently ongoing research on adaptive experiences has focused on personalization algorithms, explainability, user engagement, or privacy and security, there is growing interest and resources in developing and improving these research focuses. Even though the research on adaptive experiences has been dynamic and rapidly evolving, achieving a high level of user engagement in adaptive experiences remains a challenge. %????? This dissertation aims to uncover ways to engage users in adaptive experiences by incorporating …

Adversarial Deep Learning And Security With A Hardware Perspective, Joseph Clements

All Dissertations

Adversarial deep learning is the field of study which analyzes deep learning in the presence of adversarial entities. This entails understanding the capabilities, objectives, and attack scenarios available to the adversary to develop defensive mechanisms and avenues of robustness available to the benign parties. Understanding this facet of deep learning helps us improve the safety of the deep learning systems against external threats from adversaries. However, of equal importance, this perspective also helps the industry understand and respond to critical failures in the technology. The expectation of future success has driven significant interest in developing this technology broadly. Adversarial deep …

Explainable Physics-Informed Deep Learning For Rainfall-Runoff Modeling And Uncertainty Assessment Across The Continental United States, Sadegh Sadeghi Tabas

All Dissertations

Hydrologic models provide a comprehensive tool to calibrate streamflow response to environmental variables. Various hydrologic modeling approaches, ranging from physically based to conceptual to entirely data-driven models, have been widely used for hydrologic simulation. During the recent years, however, Deep Learning (DL), a new generation of Machine Learning (ML), has transformed hydrologic simulation research to a new direction. DL methods have recently proposed for rainfall-runoff modeling that complement both distributed and conceptual hydrologic models, particularly in a catchment where data to support a process-based model is scared and limited.

This dissertation investigated the applicability of two advanced probabilistic physics-informed DL …

Beyond Just Money Transactions: Redesigning Digital Peer-To-Peer Payments For Social Connections, Lingyuan Li

All Dissertations

Financial activities, such as the exchange of money between individuals, have long been considered a crucial aspect of how people build and maintain their interpersonal relationships (i.e., a strong, deep, or close association/acquaintance between two or more people) with individuals they know because money is a sensitive social construct. In particular, over the past decade, how to conduct, manage, and experience money exchanges and processes between individuals has been dramatically transformed due to the increasing popularity of digital peer-to-peer (P2P) payment services (i.e., performing one to one online money transactions via a digital device). In this sense, digital P2P payments …

Enhanced Mobile Networking Using Multi-Connectivity And Packet Duplication In Next-Generation Cellular Networks, Prabodh Mishra

All Dissertations

Modern cellular communication systems need to handle an enormous number of users and large amounts of data, including both users as well as system-oriented data. 5G is the fifth-generation mobile network and a new global wireless standard that follows 4G/LTE networks. The uptake of 5G is expected to be faster than any previous cellular generation, with high expectations of its future impact on the global economy. The next-generation 5G networks are designed to be flexible enough to adapt to modern use cases and be highly modular such that operators would have the flexibility to provide selective features based on user …

Vanet Applications Under Loss Scenarios & Evolving Wireless Technology, Adil Alsuhaim

All Dissertations

In this work we study the impact of wireless network impairment on the performance of VANET applications such as Cooperative Adaptive Cruise Control (CACC), and other VANET applications that periodically broadcast messages. We also study the future of VANET application in light of the evolution of radio access technologies (RAT) that are used to exchange messages. Previous work in the literature proposed fallback strategies that utilizes on-board sensors to recover in case of wireless network impairment, those methods assume a fixed time headway value, and do not achieve string stability. In this work, we study the string stability of a …

Enabling High Throughput And Reliable Low Latency Communication Over Vehicular Mobility In Next-Generation Cellular Networks, Snigdhaswin Kar

All Dissertations

The fifth-generation (5G) networks and beyond need paradigm shifts to realize the exponentially increasing demands of next-generation services for high throughputs, low latencies, and reliable communication under various mobility scenarios. However, these promising features have critical gaps that need to be filled before they can be fully implemented for mobile applications in complex environments like smart cities. Although the sub-6 GHz bands can provide reliable and larger coverage, they cannot provide high data rates with low latencies due to a scarcity of spectrum available in these bands. Millimeter wave (mmWave) communication is a key enabler for a significant increase in …

Deep Reinforcement Learning And Game Theoretic Monte Carlo Decision Process For Safe And Efficient Lane Change Maneuver And Speed Management, Shahab Karimi

All Dissertations

Predicting the states of the surrounding traffic is one of the major problems in automated driving. Maneuvers such as lane change, merge, and exit management could pose challenges in the absence of intervehicular communication and can benefit from driver behavior prediction. Predicting the motion of surrounding vehicles and trajectory planning need to be computationally efficient for real-time implementation. This dissertation presents a decision process model for real-time automated lane change and speed management in highway and urban traffic. In lane change and merge maneuvers, it is important to know how neighboring vehicles will act in the imminent future. Human driver …

Modeling, Control And Estimation Of Reconfigurable Cable Driven Parallel Robots, Adhiti Raman Thothathri

All Dissertations

The motivation for this thesis was to develop a cable-driven parallel robot (CDPR) as part of a two-part robotic device for concrete 3D printing. This research addresses specific research questions in this domain, chiefly, to present advantages offered by the addition of kinematic redundancies to CDPRs. Due to the natural actuation redundancy present in a fully constrained CDPR, the addition of internal mobility offers complex challenges in modeling and control that are not often encountered in literature.

This work presents a systematic analysis of modeling such kinematic redundancies through the application of reciprocal screw theory (RST) and Lie algebra while …

The Importance Of Hand Motions For Communication And Interaction In Virtual Reality, Alex Adkins

All Dissertations

Virtual reality (VR) is a growing method of communication and play. Recent advances have enabled hand-tracking technologies for consumer VR headsets, allowing virtual hands to mimic a user's real hand movements in real-time. A growing number of users now utilize hand-tracking when using VR to manipulate objects or to create gestures when interacting with others. As VR grows as a tool and communication platform, it is important to understand how the rising prevalence of hand-tracking technology might affect users' experiences.

The goal of this dissertation is to investigate, through a series of experiments, how using hand motions in VR influences …

Large Genomes Assembly Using Mapreduce Framework, Yuehua Zhang

All Dissertations

Knowing the genome sequence of an organism is the essential step toward understanding its genomic and genetic characteristics. Currently, whole genome shotgun (WGS) sequencing is the most widely used genome sequencing technique to determine the entire DNA sequence of an organism. Recent advances in next-generation sequencing (NGS) techniques have enabled biologists to generate large DNA sequences in a high-throughput and low-cost way. However, the assembly of NGS reads faces significant challenges due to short reads and an enormously high volume of data. Despite recent progress in genome assembly, current NGS assemblers cannot generate high-quality results or efficiently handle large genomes …

Scalable Data-Driven Predictive Modeling And Analytics For Cho Process Development Optimization, Sarah Mbiki

All Dissertations

In 1982, the FDA approved the first recombinant therapeutic protein, and since then, the biopharmaceutical industry has continued to develop innovative and highly effective biological drugs for various illnesses¹. These drugs are produced using host organisms that are modified to hold the genetic encoding of the targeted protein¹. Of the many host organisms, Chinese hamster ovary (CHO) cells are often used due to capability to perform posttranslational modification (PTM): which allows human-like synthesis of proteins unlikely to invoke immunogenicity in humans ^1,2.

Despite all the positive attributes, many challenges are associated with CHO cell cultures, …

Multi-Robot Symbolic Task And Motion Planning Leveraging Human Trust Models: Theory And Applications, Huanfei Zheng

All Dissertations

Multi-robot systems (MRS) can accomplish more complex tasks with two or more robots and have produced a broad set of applications. The presence of a human operator in an MRS can guarantee the safety of the task performing, but the human operators can be subject to heavier stress and cognitive workload in collaboration with the MRS than the single robot. It is significant for the MRS to have the provable correct task and motion planning solution for a complex task. That can reduce the human workload during supervising the task and improve the reliability of human-MRS collaboration. This dissertation relies …

Algorithm Optimization And Hardware Acceleration For Machine Learning Applications On Low-Energy Systems, Jianchi Sun

All Dissertations

Machine learning (ML) has been extensively employed for strategy optimization, decision making, data classification, etc. While ML shows great triumph in its application field, the increasing complexity of the learning models introduces neoteric challenges to the ML system designs. On the one hand, the applications of ML on resource-restricted terminals, like mobile computing and IoT devices, are prevented by the high computational complexity and memory requirement. On the other hand, the massive parameter quantity for the modern ML models appends extra demands on the system's I/O speed and memory size. This dissertation investigates feasible solutions for those challenges with software-hardware …

Snap : A Software-Defined & Named-Data Oriented Publish-Subscribe Framework For Emerging Wireless Application Systems, Manveen Kaur

All Dissertations

The evolution of Cyber-Physical Systems (CPSs) has given rise to an emergent class of CPSs defined by ad-hoc wireless connectivity, mobility, and resource constraints in computation, memory, communications, and battery power. These systems are expected to fulfill essential roles in critical infrastructure sectors. Vehicular Ad-Hoc Network (VANET) and a swarm of Unmanned Aerial Vehicles (UAV swarm) are examples of such systems. The significant utility of these systems, coupled with their economic viability, is a crucial indicator of their anticipated growth in the future. Typically, the tasks assigned to these systems have strict Quality-of-Service (QoS) requirements and require sensing, perception, and …

The Development Of Tigra: A Zero Latency Interface For Accelerator Communication In Risc-V Processors, Wesley Brad Green

All Dissertations

Field programmable gate arrays (FPGA) give developers the ability to design application specific hardware by means of software, providing a method of accelerating algorithms with higher power efficiency when compared to CPU or GPU accelerated applications. FPGA accelerated applications tend to follow either a loosely coupled or tightly coupled design. Loosely coupled designs often use OpenCL to utilize the FPGA as an accelerator much like a GPU, which provides a simplifed design flow with the trade-off of increased overhead and latency due to bus communication. Tightly coupled designs modify an existing CPU to introduce instruction set extensions to provide a …

Digitalization Of Construction Project Requirements Using Natural Language Processing (Nlp) Techniques, Fahad Ul Hassan

All Dissertations

Contract documents are a critical legal component of a construction project that specify all wishes and expectations of the owner toward the design, construction, and handover of a project. A single contract package, especially of a design-build (DB) project, comprises hundreds of documents including thousands of requirements. Precise comprehension and management of the requirements are critical to ensure that all important explicit and implicit requirements of the project scope are captured, managed, and completed. Since requirements are mainly written in a natural human language, the current manual methods impose a significant burden on practitioners to process and restructure them into …

Control, Decision-Making, And Learning Approaches For Connected And Autonomous Driving Systems With Humans-In-The-Loop, Fangjian Li

All Dissertations

By virtue of vehicular connectivity and automation, the vehicle becomes increasingly intelligent and self-driving capable. However, no matter what automation level the vehicle can achieve, humans will still be in the loop despite their roles. First, considering the manual driving car as a disturbance to the connected and autonomous vehicles (CAVs), a novel string stability is proposed for mixed traffic platoons consisting of both autonomous and manual driving cars to guarantee acceptable motion fluctuation and platoon safety. Furthermore, humans are naturally considered as the rider in the passenger vehicle. A human-centered cooperative adaptive cruise control (CACC) is designed to improve …

A Youthful Metaverse: Towards Designing Safe, Equitable, And Emotionally Fulfilling Social Virtual Reality Spaces For Younger Users, Divine Maloney

All Dissertations

Social virtual reality (VR) represents the modern rendition of the metaverse, this dissertation aims to fill the research gaps while highlighting trends of youth in VR. The scientific contributions of this research include 1) expanding the current HCI understanding of the social dynamics and the interactions of teens in emerging novel online digital spaces; 2) bridging two research areas that have not been widely studied in HCI, social VR and young users in social VR; and 3) generating design implications to inform the design of future social VR platforms to better support and protect teens’ online social experiences, results which …

Deep Learning Based Speech Enhancement And Its Application To Speech Recognition, Ju Lin

All Dissertations

Speech enhancement is the task that aims to improve the quality and the intelligibility of a speech signal that is degraded by ambient noise and room reverberation. Speech enhancement algorithms are used extensively in many audio- and communication systems, including mobile handsets, speech recognition, speaker verification systems and hearing aids. Recently, deep learning has achieved great success in many applications, such as computer vision, nature language processing and speech recognition. Speech enhancement methods have been introduced that use deep-learning techniques, as these techniques are capable of learning complex hierarchical functions using large-scale training data. This dissertation investigates the deep learning …

Optimizing Virtual Resource Management In Cloud Datacenters, Liuhua Chen

All Dissertations

Datacenter clouds (e.g., Microsoft's Azure, Google's App Engine, and Amazon's EC2) are emerging as a popular infrastructure for computing and storage due to their high scalability and elasticity. More and more companies and organizations shift their services (e.g., online social networks, Dropbox file hosting) to clouds to avoid large capital expenditures. Cloud systems employ virtualization technology to provide resources in physical machines (PMs) in the form of virtual machines (VMs). Users create VMs deployed on the cloud and each VM consumes resources (e.g., CPU, memory and bandwidth) from its host PM. Cloud providers supply services by signing Service Level Agreement …

Design Of A Wide Area Controller Using Eigenstructure Assignment In Power Systems, Parimal Saraf

All Dissertations

Small signal stability has become a major concern for power system operators around the world. This has resulted from constantly evolving changes in the power system ranging from increased number of interconnections to ever increasing demand of power. In highly stressed operating conditions, even a small disturbance such as a load change can make the system unstable resulting in small signal instability. The main reason for small signal instability in power systems is an inter-area mode/s becoming unstable. Inter-area modes involve a group of generators oscillating against each other. Traditionally, power system stabilizers installed on the synchrous machines were used …

A Time-Efficient Strategy For Relay Selection And Link Scheduling In Wireless Communication Networks, Chenxi Qiu

All Dissertations

Despite the unprecedented success and proliferation of wireless communication, sustainable reliability and stability among wireless users are still considered important issues in the underlying link protocols. Existing link-layer protocols, like ARQ [44] or HARQ [57,67] approaches are designed to achieve this goal by discarding a corrupted packet at the receiver and performing one or more retransmissions until the packet is successfully decoded or a maximum number of retransmission attempts is reached. These strategies suffer from degradation of throughput and overall system instability since packets need to be en/decode in every hop, leading to high burden for relay nodes especially when …

Cloud-Based Strategies For Robust Connectivity And Efficient Transmission In Vehicular Networks, Ke Xu

All Dissertations

Leveraging multiple wireless technologies and radio access networks, vehicles on the move have the potential to get ubiquitous broadband Internet connectivity. Many studies have put lots of efforts on vehicle-to-vehicle networks for relaying strategies, popular content distribution, etc. However, in dominant infrastructure-based vehicular networks, supporting continuous and fast data transfer for today's prevalent services, e.g. video streaming, for vehicles anytime and anywhere is still a difficult research problem. By looking into such problem, impacts such as intermittent connectivity, dynamic network topology, fluctuating signal coverage, and inefficient transmissions, all result from two root causes in vehicular network- mobility and limited infrastructure …

Towards Efficient File Sharing And Packet Routing In Mobile Opportunistic Networks, Kang Chen

All Dissertations

With the increasing popularity of portable digital devices (e.g., smartphones, laptops, and tablets), mobile opportunistic networks (MONs) [40, 90] consisting of portable devices have attracted much attention recently. MONs are also known as pocket switched networks (PSNs) [52]. MONs can be regarded as a special form of mobile ad hoc networks (MANETs) [7] or delay tolerant networks (DTNs) [35, 56]. In such networks, mobile nodes (devices) move continuously and meet opportunistically. Two mobile nodes can communicate with each other only when they are within the communication range of each other in a peer-to-peer (P2P) manner (i.e., without the need of …

Large Scale 3d Mapping Of Indoor Environments Using A Handheld Rgbd Camera, Brian Peasley

All Dissertations

The goal of this research is to investigate the problem of reconstructing a 3D representation of an environment, of arbitrary size, using a handheld color and depth (RGBD) sensor. The focus of this dissertation is to examine four of the underlying subproblems to this system: camera tracking, loop closure, data storage, and integration. First, a system for 3D reconstruction of large indoor planar environments with data captured from an RGBD sensor mounted on a mobile robotic platform is presented. An algorithm for constructing nearly drift-free 3D occupancy grids of large indoor environments in an online manner is also presented. This …

Exploring Multiple Levels Of Performance Modeling For Heterogeneous Systems, Venkittaraman Vivek Pallipuram Krishnamani

All Dissertations

The current trend in High-Performance Computing (HPC) is to extract concurrency from clusters that include heterogeneous resources such as General Purpose Graphical Processing Units (GPGPUs) and Field Programmable Gate Array (FPGAs). Although these heterogeneous systems can provide substantial performance for massively parallel applications, much of the available computing resources are often under-utilized due to inefficient application mapping, load balancing, and tuning. While several performance prediction models exist to efficiently tune applications, they often require significant computing architecture knowledge for reliable prediction. In addition, they do not address multiple levels of design space abstraction and it is often difficult to choose …

Hierarchical Channel-Access And Routing Protocols For Heterogeneous Multichannel Ad Hoc Networks, Crystal Jackson

All Dissertations

In this work, we consider a heterogeneous multichannel ad hoc network consisting of frequency-agile radios with the ability to change their carrier frequency and chip rate over a wide range of possibilities. The radios in this type of network are permitted to utilize multiple non-overlapping channels, and each channel differs significantly in its characteristics such as achievable data rate and communication range. A channel-access protocol regulates access to the channels available, and a routing protocol determines how a packet should be forwarded from its source to its destination through the network. The initial focus of this research is on channel-access …

Mobile Robot Navigation For Person Following In Indoor Environments, Ninad Pradhan

All Dissertations

Service robotics is a rapidly growing area of interest in robotics research. Service robots inhabit human-populated environments and carry out specific tasks. The goal of this dissertation is to develop a service robot capable of following a human leader around populated indoor environments. A classification system for person followers is proposed such that it clearly defines the expected interaction between the leader and the robotic follower. In populated environments, the robot needs to be able to detect and identify its leader and track the leader through occlusions, a common characteristic of populated spaces. An appearance-based person descriptor, which augments the …

Occlusion-Aware Multi-View Reconstruction Of Articulated Objects For Manipulation, Xiaoxia Huang

All Dissertations

The goal of this research is to develop algorithms using multiple views to automatically recover complete 3D models of articulated objects in unstructured environments and thereby enable a robotic system to facilitate further manipulation of those objects. First, an algorithm called Procrustes-Lo-RANSAC (PLR) is presented. Structure-from-motion techniques are used to capture 3D point cloud models of an articulated object in two different configurations. Procrustes analysis, combined with a locally optimized RANSAC sampling strategy, facilitates a straightforward geometric approach to recovering the joint axes, as well as classifying them automatically as either revolute or prismatic. The algorithm does not require prior …