University of South Alabama Institutional Repository
University of South Alabama Institutional RepositoryNot a member yet
5383 research outputs found
Sort by
Nonlinear Control System Design and Hardware Testing of an Overactuated Multicopter
Full text linkFor the concept of Urban Air Mobility to be implemented into everyday life, the safety of passengers, bystanders, and surrounding infrastructure needs to be held paramount [1]. In previous work done at the University of South Alabama Facility for Aerospace Systems and Technology (FAST) lab, a reconfigurable control law was derived and tested on a scratch built multicopter [2]. The goal of the control law was to stabilize an X8 model multicopter in the event of motor outages [2]. The control law was tested on a simulated model and flight tested on a small-scale X8 multicopter to verify the effectiveness of the control law [2]. In this thesis, the isolated software previously tested was integrated into the FAST Configurable Autopilot and Simulation Software Tool (FASTCASST) software and simulated and flight tested to verify the integration of the control law and aerodynamic model into the FASTCASST ecosystem. Simulations observing the response of the roll and pitch are observed to verify the integration of the aerodynamic model, and simulations observing the altitude, roll angle, pitch angle, and yaw rate are observed to verify the integration and effectiveness of the control law. Finally, a multicopter was flight tested to verify the integration and effectiveness of the control law into the FASTCASST ecosystem
Queer Hope and Asexual desire in Nella Larsen\u27s Passing and Virginia Woolf\u27s To the Lighthouse
Full text linkThis project examines Nella Larsen’s Passing and Virginia Woolf’s To the Lighthouse together as queer artifacts in hopes of shedding new light on how race and sexuality can be reexamined across national modes of thinking. Although these writers’ backgrounds provide them with different bodies of knowledge, Larsen’s Irene and Woolf’s Lily have much more in common than might first appear. For one, even if racialization is more explicit—and presents more of an obstacle—for Irene, race is an important factor for both characters. Additionally, Irene and Lily both express profound discontent with marriage, Irene in her marriage to Brian and Lily against Mrs. Ramsay’s insistence that she marry. In their unhappiness, however, they look toward hopeful queer futures in which they need not be sexual beings, thereby opening a window for the recent sociological field of asexuality studies. Seeking to raise awareness and encourage acceptance of asexuality as a sexual identity, asexuality studies asks us to rethink our understandings of sexuality and expand the existing body of queer studies. My goal with this project is twofold: to promote more inclusions into the field of queer studies, such as asexuality studies and considerations of race, and to revisit works of twentieth-century literature in hopes of generating a new conversation around their contemporary relevance
Research in the Age of AI
No full textA guide for higher education professionals on conducting and supporting research effectively in a rapidly changing technological environment using AI. This strategy emphasizes the importance of critical engagement with research sources, understanding the cultural impact of AI technology, and using AI tools responsibly and strategically. It outlines a framework for navigating research in the age of AI, focusing on controlling the corpus of information, engaging in critical thinking, and understanding the various roles of AI in research processes. The strategy advocates for continuous learning and adaptation to technological change, emphasizing the need to approach AI tools with ethical considerations, privacy concerns, and a clear understanding of their purpose and potential benefits
Advancing Sustainable Road Construction with Cold Mix RAP and Rejuvenator
Full text linkPresentation slides for a presentation given at the 1st annual Shelby Hall Graduate Research Forum at the University of South Alabama
Directing Sophisticated Cyberattacks on Public Water Infrastructure
No full textIn recent years there has been an increasing number of cyberattacks on public water generation and distribution systems. Advanced persistent attackers could usurp sensor and control systems to contaminate public drinking water. In order to conceal their malicious activity, they can manipulate sensor data flows to give the appearance of normal activity. The compromised sensors would report normal chemical levels even though unsafe water is entering the distribution system. In response, this research proposes a multi-sensor, cross-validation approach to anomaly detection. The proposed approach is designed to detect sophisticated cyberattacks which are not easily detectable using traditional cyber tools. The proposed solution would leverage the interrelationship between secondary measures and safe consumable water metrics to detect attacks on water distribution systems. It could be employed to detect attacks on water supplies in near real-time operating environments.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1002/thumbnail.jp
Edge AI-Based Cognitive Engagement Detection and Multimodal Exploration
No full textEdge AI models for real-time cognitive engagement detection will have major effects on clinical neuroscience, human-computer interface, and education. This work provides a lightweight deep learning model for frame-by- frame video classification and develops an edge device. The proposed edge device delivers low latency, energy-efficient processing, and high temporal resolution, achieving 29 frames per second (FPS) with each frame taking 30 milliseconds, all without reliance on cloud services. Our investigation also introduces a Multimodal Fusion Framework designed to integrate physiological signals and behavioral data into the engagement classification process, in contrast to conventional systems that rely solely on video-based categorization. Using synchronized EEG, ECG, and face video data from 17 student volunteers engaged in controlled cognitive activities, the study examined. While EEG alpha and beta frequency bands gave important new perspectives on cognitive workload, ECG-derived heart rate variability (HRV) acted as an indirect predictor of cognitive stress. Reaction times and task accuracy among behavioral measures provided outside confirmation of cognitive states. This work investigates the strengths and limits of each modality by means of independent and integrated data streams, therefore stressing the possibilities of improving classification accuracy and contextual knowledge of engagement levels. In contrast to conventional systems that mostly rely on video-based categorization, this study provides a Multimodal Fusion Framework that integrates behavioral and physiological information into the engagement classification process. Without requiring cloud infrastructure, the proposed edge device demonstrates low latency, energy-efficient computation, and excellent temporal resolution.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1010/thumbnail.jp
Low-Power Differential Privacy Memory for Edge Devices
No full textAs the use of personal electronics have increased, new and improved services and applications have become the cornerstone of modern times. To improve these services, companies and institutions collect data to inform new features that are needed to support their services. Differential Privacy is a methodology to ensure that this collected data does not leak any individual’s personal information while still allowing data collectors to see trends in data. In this work, Differential Privacy is implemented at a hardware-level by injecting noise in memory. To design this memory, the failure rate of SRAM cells at different voltage levels was evaluated and used to determine several device parameters.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1016/thumbnail.jp
Low-Cost Wearable Edge-AI Device for Diabetes Management
No full textDiabetes Mellitus (DM) refers to a set of serious metabolic disorders affecting 537 million people worldwide, according to 2021 statistics from the International Diabetes Federation. Blood glucose (BG) levels are a vital health indicator for managing DM, but conventional methods of measuring BG use direct blood or interstitial fluid samples. While accurate, these ‘invasive’ methods can cause discomfort, tissue damage, pose a risk of infection for the patient, and often require expensive consumable materials or devices to use. Developing a ‘noninvasive’ method of measuring or approximating BG has grown increasingly relevant as healthcare technology advances and cases of DM continue to rise. Artificial intelligence (AI) techniques have shown promising results in BG prediction using a variety of noninvasively sampled signals. However, these techniques are seldom implemented on wearable edge devices owing to complex sensors, costly algorithms, and reliance on cloud computing. This work designs, prototypes, and evaluates a standalone wearable device to estimate BG with AI methods, with emphasis placed on integration of software, hardware, and user interaction. Static information from user input and a photoplethysmography sensor are used to provide noninvasive data for the AI model of this device to estimate BG. The device achieves a prediction accuracy of 16.8% mean average percentage error and produces mostly clinically acceptable results according to Clarke Error Grid Analysis.https://jagworks.southalabama.edu/southalabama-shgrf-posters/1019/thumbnail.jp
Comparative Analysis of Classical and Machine Learning Pathfinding Approaches
Full text linkPathfinding is an essential task for any autonomous robot. Graph-based classical pathfinding algorithms and machine learning approaches have both been used for this end, but they are often not compared against each other. An implementation of end-to-end (E2E) pathfinding using Proximal Policy Optimization (PPO) and an Alexnet architecture is compared against an implementation of Hybrid A*. A digital twin in Unity3D is used as the testing environment with the Clearpath Dingo as the pathfinding robot. In machine learning, the robot is controlled using PPO through ROS-Noetic with a camera as its sensor. Hybrid A* and its controls are implemented directly in Unity3D. The two pathfinding algorithms were evaluated on path length and traversal time along a marked route in simulation. They are also evaluated on training time, path calculation time, and storage space. Experimental results showed that PPO had a success rate of only 28.68% over 10801 test runs. In successful runs, PPO produced paths that were 13.91 meters on average with a standard deviation of ± 0.22 meters, traversed in 34.32 seconds on average with a standard deviation of ± 4.54 seconds. Hybrid A* produced a path that was 13.36 meters, traversed in 22.69 seconds
TOUJOURS BEAUCOUP: Representations of Childhood and Education in 20th Century French Literature and Cinema
Full text linkThe Third Republic’s reforms established France’s public education system on several foundational ideals: upholding republican values, resisting autocratic authorities, promoting critical thinking, and promoting personal liberty and freedom of thought. This project explores French literature and cinema with significant representations of education for the purpose of examining to what extent such ideals are upheld or undermined in their portrayals of education. Such an examination of French education and its underlying values is significant in such a time in history when France is experiencing a “contemporary crisis of republican schooling” (Sachs 179). Alongside the historical significance of such studies, the more general need for this kind of research is humanity’s social responsibility for preparing the next generation to take their place in the forefront of civilization – naturally, this involves ensuring that a society’s education system and general approaches to the development of its children are of the highest possible quality. Chapter 1 will explore how select French works portray the ideals of French education, especially the promise of how republican education can positively shape children’s lives. Chapter 2 will then explore how representations of French education expose the education system for, in many ways, failing to uphold its ideals and even negatively impacting students’ lives. The conclusion will then synthesize these findings, proposing a well-rounded understanding of French education as a system that falls short of its values in many ways, but which maintains potential in returning to its foundational ideals