31825 research outputs found
Sort by
Leveraging deep learning for robust visual and visual-inertial SLAM
No full textHuang, GuoquanSimultaneous localization and mapping (SLAM) arises in numerous applications ranging from robot navigation to augmented reality (AR) and virtual reality (VR). Despite recent advances in deep learning, most SLAM systems still rely completely on hand-crafted techniques. In this dissertation, we aim to show how deep learning can benefit SLAM -- in certain cases improving the accuracy, robustness, and even sometimes the efficiency. We show not only that deep learning can benefit SLAM, but in many cases SLAM can actually benefit the deep neural network in return, exposing a mutually-beneficial relationship between the learned and hand-crafted techniques. The dissertation is divided in four thrusts. In the first thrust, a robust visual object SLAM system is presented, which utilizes a custom deep semantic keypoint network to provide the monocular SLAM system with metric scale while allowing it to initialize from a single view. In return, the SLAM system provides valuable priors to the network which allows it to track keypoints on symmetric objects consistently across multiple views, which is typically not possible. In the second thrust, a dense visual-inertial odometry (VIO) system is presented, in which dense geometry is represented as a compact optimizable code that is estimated tightly in the VIO estimator. We show that not only the code estimation can improve the dense geometry's accuracy, but that it can also improve the accuracy of VIO in return. In the third thrust, a monocular depth-aided visual-inertial initialization pipeline is presented. The learned monocular depth is shown to improve the initialization performance in low-excitation scenarios where completely hand-crafted initialization performance is degraded, and in turn the inertial information provides the scale to the monocular depth network which allows it to be used as a compact feature representation. In the fourth and final thrust, an efficient and robust dense visual-inertial SLAM system is presented called AB-VINS. Instead of estimating each feature position separately, as in most SLAM systems, AB-VINS utilizes a monocular depth network to represent the geometry, which is shown to improve the robustness and efficiency. Again, the monocular depth provides useful priors to the SLAM system while the inertial information provides the scale to the network. A new hand-crafted technique for pose graph optimization called the memory tree is introduced along with AB-VINS, which is shown to greatly improve the robustness and efficiency over state-of-the-art methods -- allowing for AB-VINS to solve pose graph SLAM while only relinearizing a constant number of variables no matter the number of keyframes. The dissertation is concluded with possible future research directions.University of Delaware, Department of Computer and Information SciencesPh.D
Why do students choose the option of the Open Syllabus in a conventional university?
No full textThis article was originally published in Dialogic Pedagogy. The version of record is available at: https://doi.org/10.5195/dpj.2025.695.
This work is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).The purpose of the presented mixed qualitative-quantitative research is to examine college students’ diverse reasons for choosing the Open Syllabus, which allows students in a conventional university to define their goals for education, curriculum, instruction, assessment, ways of learning, and so on—what traditionally constitutes “Self-Directed Education.” Most of those students articulated their interest in self-education, which consists of self-directed and responsive education
2025 KIDS COUNT in Delaware FOCUS on Strength in Numbers
No full textSimilar to previous editions of our annual FOCUS Book, the 2025 KIDS COUNT in Delaware FOCUS on Public Policy provides an overview of child well-being in Delaware and reports on select indicators across the four traditionally reported categories of health, education, economic security, and family/community. This year's 30th anniversary edition of the FOCUS book reflects on how far we’ve come—and how data can guide us toward a better future for Delaware’s children and families. We’ll explore long-term trends, spotlight persistent challenges, and highlight how local leaders, policymakers, and advocates can use this information to drive change.Funded by The Annie E. Casey Foundation, the University of Delaware and the State of Delaware
Impact of disasters on adverse pregnancy outcomes: insights from the Environmental Influences on Child Health Outcomes (ECHO) Program
No full textHorney, JenniferAs disasters grow in frequency and severity across the United States, there is increasing concern about their impact on maternal and child health. Pregnant individuals represent a uniquely vulnerable population, but few epidemiologic studies have rigorously examined how disaster exposure influences adverse pregnancy outcomes. Limitations in prior research include small, non-diverse samples; inconsistent measurement of disaster exposure; and inadequate attention to the role of timing, frequency, severity, and social determinants of health. This dissertation leverages harmonized, individual-level data from the NIH-funded Environmental Influences on Child Health Outcomes (ECHO) Program to evaluate disaster-related pregnancy risks in a large, diverse cohort. ☐ This dissertation aimed to: (1) characterize sociodemographic and health differences among disaster-exposed and unexposed pregnant individuals; (2) estimate the association between disaster exposure and five key adverse pregnancy outcomes—gestational hypertension (GHTN), gestational diabetes mellitus (GDM), preterm birth (PTB), small for gestational age (SGA), and large for gestational age (LGA); (3) examine whether associations differ by race and ethnicity, education, and income; and (4) explore the influence of disaster timing, frequency, and severity on pregnancy risks. ☐ A retrospective cohort study was conducted using data from 3,142 participants across 14 cohorts within the ECHO Release 2 dataset. Disaster exposure was self-reported using the Life Stressor Checklist-Revised and operationalized in binary, frequency, severity, and timing formats. Modified Poisson regression with generalized estimating equations (GEE) was used to estimate relative risks (RR) for each outcome, accounting for cohort-level clustering. Multiple imputation addressed missing covariate data. Stratified models and interaction terms tested effect modification. Sensitivity analyses included complete case models, leave-one-out cohort exclusion, and non-parametric Kaplan–Meier–like survival curves using maternal age as the time scale. ☐ Disaster exposure was reported by 48% of participants. Exposed individuals were more likely to have lower income and education levels, identify as Black or Hispanic, and report higher rates of asthma, depression, and interpersonal violence. These findings underscore the intersection of disaster vulnerability and pre-existing social inequities. ☐ While disaster exposure showed modest associations with reduced risk of GHTN (RR = 0.59; 95% CI: 0.29–1.07) and PTB (RR = 0.84; 95% CI: 0.70–1.00), the magnitude and direction of associations varied across subgroups. For example, protective associations for GHTN were more pronounced among low-income individuals and those with lower educational attainment. No strong associations were found for GDM, SGA, or LGA, and effect modification by race or income was generally weak or inconsistent for these outcomes. ☐ Recent disaster exposure (within five years of conception) was associated with reduced risk of GHTN and PTB, while disaster severity was positively associated with GHTN. Exposure frequency did not show a clear dose-response relationship. Kaplan–Meier–like curves illustrated outcome distributions by disaster exposure, but curve crossings suggested violations of the proportional hazards assumption, limiting interpretation of traditional Cox survival models. ☐ This dissertation provides new evidence on the nuanced relationship between disaster exposure and pregnancy outcomes. While some findings suggest a protective association between disaster exposure and select outcomes such as GHTN and PTB, these patterns may reflect underlying resilience, access to care, or selection mechanisms. Importantly, disaster exposure intersects with broader social determinants of health, highlighting the need for equity-centered disaster response and maternal health planning. Heterogeneity in associations across timing and severity of exposure supports the need for more refined disaster characterization in perinatal epidemiology. Future research should incorporate longitudinal designs and mechanistic pathways to better understand how disasters shape reproductive health across different population contexts.University of Delaware, Department of EpidemiologyPh.D
Utterance phenomena in syntax
No full textTomioka, SatoshiThis dissertation explores utterance-oriented phenomena in syntax, focusing on the integration of discourse-related elements into sentence structure. Utterance-oriented phenomena are sentence-peripheral, conversational linguistic expressions that convey expressive meaning in terms of speaker attitude. By analyzing data from Turkish and Japanese, the dissertation presents a formal syntactic framework that accommodates these phenomena within an expanded syntactic hierarchy, particularly through the functional projection of the Utterance Phrase (UttP). The empirical data representing utterance-oriented phenomena come from three different constructions in this work: utterance oriented markers, gapped right dislocation and gapless right dislocation. Utterance-oriented markers, a class of discourse particles that satisfy the criteria for being utterance-oriented phenomena, are argued to base-adjoined to UttP either at the head level or the phrasal level. Gapped right dislocation in Turkish and Japanese is taken to involve a rightward syntactic movement to UttP, where they adjoin to the phrasal level. Gapless right dislocation, which come in different forms, are associated with different syntactic structures depending on their syntactic, semantic and pragmatic properties. The dissertation overall contributes to the broader understanding of how syntactic structure interfaces with discourse, offering a nuanced view of how conversational phenomena are syntactically represented across these two languages, with a potential cross-linguistic implications.University of Delaware, Department of Linguistics and Cognitive SciencePh.D
Decentralizing energy: prosumers, carbon incentives, and market behaviors
No full textShah, KalimThis dissertation investigates the behavioral, economic, and institutional dimensions of prosumer-centered transitions in peer-to-peer (P2P) energy trading systems. Through four interlinked studies, it provides a comprehensive exploration of how decentralized energy platforms can be designed to support household-level energy autonomy while advancing market efficiency and climate mitigation goals. The research is situated in the context of Mauritius, a small island developing state facing pressing challenges of energy security, affordability, and decarbonization, and combines empirical modeling, survey experimentation, and market design. ☐ Chapter 2 provides a comprehensive review of P2P electricity trading. It synthesizes current academic and practical developments in market design, pricing mechanisms, and enabling technologies. The review categorizes market designs into centralized, decentralized, and distributed structures, and compares four pricing approaches: optimization, game theory, auction-based systems, and reinforcement learning. It highlights how emerging technologies like blockchain, IoT, artificial intelligence, and advanced communication networks support P2P energy systems. Importantly, the paper identifies persistent challenges in social acceptance, scalability, and regulation, especially the social ontology of trading behavior. A STEP analysis (Social, Technological, Economic, Political) frames future research needs to enhance the scalability, equity, and trustworthiness of P2P markets. This review offers critical insights for researchers and practitioners seeking to align decentralized energy markets with digital and decarbonized energy transitions. ☐ Chapter 3 explores the behavioral underpinnings of household participation in P2P energy trading, focusing on the trade-off between energy autonomy and financial incentives. Drawing on a nationwide survey with 111 households and applying multilevel regression and cluster analysis, the study identifies three distinct trading behavior archetypes: autonomy-oriented prosumers who limit trading to preserve stored energy, price-driven prosumers who prioritize economic gain, and non-monotonic prosumers with inconsistent trading patterns. The findings highlight that market liquidity in decentralized energy systems is shaped not only by physical capacity but by psychosocial preferences, underscoring the importance of designing user-centric platforms that accommodate behavioral diversity. ☐ Chapter 4 proposes a novel conceptual extension to the P2P trading model – the P2P+C framework –which integrates carbon offset monetization into local energy exchange. The model builds on the premise that P2P trading displaces grid-based fossil fuel consumption, generating verifiable carbon savings that can be bundled and sold on voluntary or compliance carbon markets. This integration creates a dual revenue stream for prosumers: earnings from electricity sales and dividends from carbon credit revenue. The framework positions P2P+C as a tool to link local energy resilience with global climate finance, particularly in off-grid and underserved regions, thereby democratizing access to sustainability resources and enabling financial transfers from carbon buyers in the Global North to communities in the Global South. ☐ Chapter 5 empirically tests the behavioral impact of the P2P+C model using a within-subjects Difference-in-Differences (DiD) design embedded in multilevel regression analysis. The results show that introducing carbon revenue significantly enhances both willingness to trade and price responsiveness among prosumers. Carbon incentives were especially effective in shifting the behavior of autonomy-oriented and threshold-based participants, suggesting that environmental monetization can act as a behavioral modulator in energy decision-making. Importantly, the study finds that prosumer engagement is better predicted by traits like risk-taking and community orientation than by standard socio-demographic factors, signaling the need for more nuanced segmentation and incentive strategies in P2P system design. ☐ Taken together, these chapters demonstrate that P2P energy trading systems are not purely technological constructs but socio-technical ecosystems shaped by behavioral dynamics and policy frameworks. The dissertation makes three central contributions: (1) it unpacks the micro-level behavioral constraints that influence market-level outcomes; (2) it introduces a financially viable mechanism for integrating climate value into decentralized energy systems; and (3) it empirically validates the effectiveness of compound incentive structures in enhancing participation. The findings offer actionable insights for platform developers, energy regulators, and climate finance institutions aiming to scale decentralized, equitable, and climate-aligned energy systems.University of Delaware, School of Public Policy and AdministrationPh.D.University of Delaware, Energy and Environmental Policy Progra
Force-Induced Selective Carbon-Carbon Bond Cleavage inMechanoresponsive Topochemical Polymers
No full text©2025 The Author(s). Advanced Materials published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/)
which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
This article was originally published in Advanced Materials. The version of record is available at: https://doi.org/10.1002/adma.202510482Mechanoresponsive polymeric materials that respond to mechanical deformation are highly valued for their potential in sensors, degradation studies, and optoelectronics. However, direct visualization and detection of these responses remain significant obstacles. In this study, novel mechanoresponsive polybiidenedionediyl (PBIT) derivative topochemical polymers are developed that depolymerize under mechanical forces, exhibiting a distinct and irreversible color change in response to grinding, milling, and compression. This color change is attributed to the alteration of polymer backbone conjugation during elongated Carbon-Carbon (C─C) single bond cleavage. Quantum chemical pulling simulations on PBIT polymers reveals a force range of 4.3–5.0 nN associated with the selective cleavage of elongated C─C single bonds. This force range is comparable to that observed for typical homolytic mechanophores, supporting the mechanistic interpretation of homolytic bond scission under mechanical stress. C─C bond cleavage kinetic studies of PBIT under compression indicates that strong interchain interactions significantly increase the pressure needed to cleave the elongated C─C bonds. Additionally, PBIT polymer thin films are composited with polydimethylsiloxane to create free-standing and robust thin films, which can serve as ink-free and rewritable paper for writing and stress visualization applications. This advancement opens new possibilities for utilizing crystalline and brittle topochemical polymers in practical applications.This work was supported by the Davidson School of Chemical Engineering
of Purdue University and American Chemical Society Petroleum Research
Fund (PRF# 68153-ND7). This work was supported by the National Supercomputing
Center with supercomputing resources including technical
support (KSC-2023-CRE-0317). The authors thank HaiyanWang, Xinghang
Zhang, and Bo Yang for their assistance with the ball milling tests of PBIT
polymers. The authors thank Tingting Gu for her valuable assistance with
visualizing the mechanical depolymerization process. The authors are also
grateful to Matthias Zeller for his support with single-crystal analysis of
the PBIT polymers. Special thanks to Yuanhao Tang for his help with largescale
fabrication of PBIT polymer thin films. And the authors thank Bryan
W. Boudouris, Jianguo Mei, and Hohyung Kang for useful discussions
Assessing public health disparities: impact of the Beltline Project on the city of Atlanta
No full textYanich, DaniloUrbanization has led to the majority of the American population residing in densely developed areas, prompting a critical examination of how limited urban space is utilized. This study investigates the importance of integrating green spaces into urban planning and development to address communities' physical and mental health needs. Drawing on 2010 and 2020 data from the US Census Bureau and the Center for Disease Control and Prevention, this research explores repurposing existing infrastructure into urban green spaces and their potential health benefits. ☐ This analysis aimed to reveal whether urban green spaces, like the BeltLine in Atlanta, Georgia could aid in mitigating poor public health indicators, amongst the population who reside within the city. The Atlanta BeltLine, a 22-mile loop of trails, transit, and greenspace, serves as an exemplary case study due to its diverse neighborhoods and historical railway origins. Initially constructed in the late 19th century and largely abandoned by the mid-20th century, the BeltLine now aims to drive sustainable growth and connectivity. This would be achieved by the green spaces promoting physical activity and encouraging behavior that could improve mental health in residents with access to this green space. By comparing public health data collected by the U.S. Census Bureau and the Center for Disease Control and Prevention, this analysis examined differences in the public health indicators for census tracts in the city of Atlanta for those that do and do not have direct contact with the BeltLine Project. ☐ The hypothesis examines if the tracts with access and direct contact with the BeltLine Projects have better public health outcomes compared to tracts that do not. This study reveals that while all Atlanta census tracts saw health conditions worsen between 2010 and 2020, those with access to the BeltLine experienced a slower rate of decline. Although the changes for most were not statistically significant, the data indicates that green infrastructure could potentially mitigate the deterioration of public health indicators if given more time to work within the community.University of Delaware, School of Public Policy and AdministrationM.A
Assessing democracy in Ghana and South Africa: a comparative study of electoral, liberal, and participatory dimensions
No full textBauer, GretchenAs analysts point to a period of a global democratic recession, this project extends the discourse to two “beacons of democracy” in Africa—Ghana and South Africa. Using Varieties of Democracy (V-Dem) as a framework, it explores the state of democracy in these two countries comparatively, and specifically on electoral, liberal, and participatory dimensions. The democratic assessment further theorizes the strengths of Ghana and South Africa, as well as potential areas of democratic erosion. The author concludes that, upon conceptualizing democracy as a disaggregation of three components, erosion is occurring in some areas while not in others. Specifically, in both countries liberal democracy is declining due to endemic corruption and state capture, while electoral democracy remains steady and robust. The divergence comes with participatory democracy wherein Ghana’s civil society and consistent voter turnout proves beneficial, whereas South Africa’s precipitous declines in electoral participation are deleterious. The project also demonstrates the utility of qualitative research, and particularly for interrogating concepts like “democracy” which inherently contain great depth and nuance.University of Delaware, Department of Political Science and International RelationsPh.D
NETWORK ANALYSIS REVEALS CONSERVED PROTEIN CLUSTERS AND RHIZOSPHERE-DRIVEN BIOGEOGRAPHY IN BRADYRHIZOBIUM PHAGES
No full textenterBradyrhizobia are nitrogen-fixing bacteria that form symbiotic nodules on the
roots of leguminous plants, supplying their hosts with fixed nitrogen in exchange for
photosynthetic products. This symbiosis reduces the need for chemical fertilizers,
which in excess can lead to eutrophication and toxic algal blooms, ultimately
contribute to climate change. Consequently, the Bradyrhizobium-legume relationship
is recognized for its economic, ecological, and agricultural significance. However,
bacteriophages can shape this mutualistic relationship by regulating host bacterial
populations. The impact of bacteriophages on soil nutrient cycles and food web
interactions remains poorly understood, therefore contributing to the need for further
investigation into soil viral populations and their ecological roles.
In this study, we used previously isolated and sequenced Bradyrhizobium
phage (bradyphage) population representative genomes from Delaware soybean field
soil samples as references. Through metagenomic analysis, we identified uncultivated
viral genomes (UViGs) in environmental samples that share significant protein
homology with these bradyphages. Our results reveal conserved functional gene
groups, especially those involved in phage infection and replication, indicating the
essential roles of structural and DNA synthesis proteins. We also found bradyphages
to exhibit strong genomic similarity to UViGs from soil and rhizosphere
environments, suggesting adaptation to similar ecological niches. By identifying
conserved gene clusters and characterizing genome features of bradyphages, this work
provides a foundation for future research on viral-host interactions, with implications
for engineering phages to enhance nitrogen fixation in sustainable agriculture. Our
findings expand the known diversity of soil viruses and highlight the importance of
phage research in understanding and managing soil microbial communities.ente