Computational Engineering Commons^™

Estimating Animal Pose Using Deep Learning A Trained Deep Learning Model Outperforms Morphological Analysis, Sanghoon Lee, Jarod Banzon, Kevin Le, Dal Hyung Kim

Computer Sciences and Electrical Engineering Faculty Research

INTRODUCTION: Analyzing animal behavior helps researchers understand their decision-making process and helper tools are rapidly becoming an indispensable part of many interdisciplinary studies. However, researchers are often challenged to estimate animal pose because of the limitation of the tools and its vulnerability to a specific environment. Over the years, deep learning has been introduced as an alternative solution to overcome these challenges.

OBJECTIVES: This study investigates how deep learning models can be applied for the accurate prediction of animal behavior, comparing with traditional morphological analysis based on image pixels.

METHODS: Transparent Omnidirectional Locomotion Compensator (TOLC), a tracking device, is used …

Introducing Normal Forms To Students: A Comparison Of Theory-First Vs. Project-First Educational Approaches, Dakota C. Cookenmaster, Jacob A. Bahn

Campus Research Day

One of the primary challenges instructors face when educating scientists and engineers is the introduction of important theory, both to show its essential nature in research as well as its practicality. This paper analyzes two pedagogical methods for the instruction of database normal forms. The first of these methods is a theory-based approach that relies on written works and practices to introduce the concept. The second of these focuses on a project-based approach which aligns with normal form standards as students implement a database schema. This paper compares these in a small study and advises an appropriate method for educators.

Synthesizing Maritime Interaction Scenarios For Testing Autonomy, Benjamin E. Hargis, Yiannis E. Papelis

Modeling, Simulation and Visualization Student Capstone Conference

This paper presents a method to deterministically synthesize maritime traffic interactions that can be presented to a system under test regardless of the state of the system under test. A background to the problem is given and the method is briefly outlined. Results indicate that the approach can enable more robust evaluation of maritime autonomous algorithms.

Applications Of Parallel Discrete Event Simulation, Erik J. Jensen

Modeling, Simulation and Visualization Student Capstone Conference

This work presents three applications of parallel discrete event simulation (PDES), which describe the motivation for and the benefits of using PDES, the kinds of synchronization algorithms that are used, and scaling behavior with these different synchronization algorithms.

Real-Time External Labeling For Interactive Visualization In Virtual Environments, Shan Liu, Yuzhong Shen

Modeling, Simulation and Visualization Student Capstone Conference

A real-time external labeling algorithm has been developed to explore the potential for applying annotation and visualization to virtual reality environments, which manages label placement in the projections of virtual 3D models on the view plane. The approach intends to place labels with visual constraints, such as no overlapping, intersections, and occlusions, close proximity to the model parts, by adjusting external annotations' positions concerning available space in the view plane. This algorithm is based on the projected model's contour and adapts to camera viewpoint changes within interactive frame rates. It solves the visibility problem of annotations and operates in real-time …

Method Of Embedded Imperfections For The Direct Simulation Of Deformation Instabilities In Film-Substrate Structures, Siavash Nikravesh Kazeroni

Mechanical Engineering ETDs

In this dissertation, a novel finite-element methodology called “embedded imperfections” is proposed and employed for computationally simulating various types of deformation instabilities observed in film-substrate structures subjected to mechanical loading. The approach involves the incorporation of elements having distinctive material properties within the film-substrate interface. One can interpret this practice as a deliberate distribution of material defects within the numerical model. It has been shown that embedded imperfections not only can trigger the onset of instability, but also can lead to “direct” simulation of deformation instability problems in that primary and subsequent instability modes can all be captured in a …

Intra-Hour Solar Forecasting Using Cloud Dynamics Features Extracted From Ground-Based Infrared Sky Images, Guillermo Terrén-Serrano

Electrical and Computer Engineering ETDs

Due to the increasing use of photovoltaic systems, power grids are vulnerable to the projection of shadows from moving clouds. An intra-hour solar forecast provides power grids with the capability of automatically controlling the dispatch of energy, reducing the additional cost for a guaranteed, reliable supply of energy (i.e., energy storage). This dissertation introduces a novel sky imager consisting of a long-wave radiometric infrared camera and a visible light camera with a fisheye lens. The imager is mounted on a solar tracker to maintain the Sun in the center of the images throughout the day, reducing the scattering effect produced …

Three-Dimensional Numerical Modeling Of Flow Hydrodynamics And Cohesive Sediment Transport In Enid Lake, Mississippi, Xiaobo Chao, A. K. M. Azad Hossain, Mohammad Z. Al-Hamdan, Yafei Jia, James V. Cizdziel

Faculty and Student Publications

Enid Lake is one of the largest reservoirs located in Yazoo River Basin, the largest basin in the state of Mississippi. The lake was impounded by Enid Dam on the Yocona River in Yalobusha County and covers an area of 30 square kilometers. It provides significant natural and recreational resources. The soils in this region are highly erodible, resulting in a large amount of fine-grained cohesive sediment discharged into the lake. In this study, a 3D numerical model was developed to simulate the free surface hydrodynamics and transportation of cohesive sediment with a median diameter of 0.0025 to 0.003 mm …

Making In Media Education: An Activity-Oriented Approach To Digital Literacy, Thomas Knaus

Journal of Media Literacy Education Pre-Prints

Why is maker education a suitable approach for giving learners the 21^st century skills they need to cope with the digital transformation? This article provides an answer and represents a defense of maker education in the field of educational science. Taking a human-media-machine interaction model as the basis for discussion, this article highlights the growing importance of digital technology as well as technological principles for human communication and interaction. Communication technology and the influence of technology on culture and society require a broad understanding of media literacy in the sense of digital literacy.

Development Of Guidelines For Collecting Transit Ridership Data, Hong Yang, Kun Xie, Sherif Ishak, Qingyu Ma, Yang Liu

Computational Modeling & Simulation Engineering Faculty Publications

Transit ridership is a critical determinant for many transit applications such as operation optimizations and project prioritization under performance-based funding mechanisms. As a result, the quality of ridership data is of utmost importance to both transit administrative agencies and transit operators. Many transit operators in Virginia report their ridership data to the Department of Rail and Public Transportation (DRPT) and the National Transit Database (NTD). However, with no specific guidelines available to transit agencies in Virginia for collecting ridership data, the heterogeneous mixture of diverse data collection methods and technologies has often raised concerns about the consistency and quality of …

Representation Learning For Chemical Activity Predictions, Mohamed S. Ayed

Dissertations, Theses, and Capstone Projects

Computational prediction of a phenotypic response upon the chemical perturbation on a biological system plays an important role in drug discovery and many other applications. Chemical fingerprints derived from chemical structures are a widely used feature to build machine learning models. However, the fingerprints ignore the biological context, thus, they suffer from several problems such as the activity cliff and curse of dimensionality. Fundamentally, the chemical modulation of biological activities is a multi-scale process. It is the genome-wide chemical-target interactions that modulate chemical phenotypic responses. Thus, the genome-scale chemical-target interaction profile will more directly correlate with in vitro and in …

Sustainable Computing - Without The Hot Air, Noman Bashir, David Irwin, Prashant Shenoy, Abel Souza

Publications

The demand for computing is continuing to grow exponentially. This growth will translate to exponential growth in computing's energy consumption unless improvements in its energy-efficiency can outpace increases in its demand. Yet, after decades of research, further improving energy-efficiency is becoming increasingly challenging, as it is already highly optimized. As a result, at some point, increases in computing demand are likely to outpace increases in its energy-efficiency, potentially by a wide margin. Such exponential growth, if left unchecked, will position computing as a substantial contributor to global carbon emissions. While prominent technology companies have recognized the problem and sought to …

Image Geo-Localization With Cross-Attention, Connor Greenwell

Theses and Dissertations--Computer Science

The problem of estimating the location from which un-geotagged photographs were captured has been well studied by the computer vision community in recent years. The central proposal of this thesis is to define a common framework within which existing approaches can be constructed and evaluated, and to introduce a new method under this framework which uses cross-attention between the query image and a database of satellite imagery with known geotags. Our experiments fit within three broad categories: 1) evaluating the ability of image localization approaches to generalize to unseen regions; 2) examining performance changes under various reference database resolutions, scales, …

Numerical Simulation On The Fire Performance Of Reinforced Concrete Flat Slab, Elias Ali

Summer Community of Scholars (RCEU and HCR) Project Proposals

No abstract provided.

Design And Development Of Active Flow Effectors, Konstantinos Kanistras

Summer Community of Scholars (RCEU and HCR) Project Proposals

No abstract provided.

Design And Development Of An Online Wind Tunnel, Konstantinos Kanistras

Summer Community of Scholars (RCEU and HCR) Project Proposals

No abstract provided.

Simulation-Based Optimization: Implications Of Complex Adaptive Systems And Deep Uncertainty, Andreas Tolk

VMASC Publications

Within the modeling and simulation community, simulation-based optimization has often been successfully used to improve productivity and business processes. However, the increased importance of using simulation to better understand complex adaptive systems and address operations research questions characterized by deep uncertainty, such as the need for policy support within socio-technical systems, leads to the necessity to revisit the way simulation can be applied in this new area. Similar observations can be made for complex adaptive systems that constantly change their behavior, which is reflected in a continually changing solution space. Deep uncertainty describes problems with inadequate or incomplete information about …

A Literature Review On Combining Heuristics And Exact Algorithms In Combinatorial Optimization, Hesamoddin Tahami, Hengameh Fakhravar

Engineering Management & Systems Engineering Faculty Publications

There are several approaches for solving hard optimization problems. Mathematical programming techniques such as (integer) linear programming-based methods and metaheuristic approaches are two extremely effective streams for combinatorial problems. Different research streams, more or less in isolation from one another, created these two. Only several years ago, many scholars noticed the advantages and enormous potential of building hybrids of combining mathematical programming methodologies and metaheuristics. In reality, many problems can be solved much better by exploiting synergies between these approaches than by “pure” classical algorithms. The key question is how to integrate mathematical programming methods and metaheuristics to achieve such …

Combining Green Metrics And Digital Twins For Sustainability Planning And Governance Of Smart Buildings And Cities, Casey R. Corrado, Suzanne M. Delong, Emily G. Holt, Edward Y. Hua, Andreas Tolk

VMASC Publications

Creating a more sustainable world will require a coordinated effort to address the rise of social, economic, and environmental concerns resulting from the continuous growth of cities. Supporting planners with tools to address them is pivotal, and sustainability is one of the main objectives. Modeling and simulation augmenting digital twins can play an important role to implement these tools. Although various green best practices have been utilized over time and there are related attempts at measuring green success, works in the published literature tend to focus on addressing a single problem (e.g., energy efficiency), and a comprehensive approach that takes …

Improving Feature Learning Capability And Interpretability Of Unsupervised Neural Networks, Chathurika S. Wickramasinghe Brahmana

Theses and Dissertations

The motivation for this dissertation is two-prong. Firstly, the current state of machine learning imposes the need for unsupervised Machine Learning (ML). Secondly, once such models are developed, a deeper understanding of ML models is necessary for humans to adapt and use such models.

Real-world systems generate massive amounts of unlabeled data at rapid speed, limiting the usability of state-of-the-art supervised machine learning approaches. Further, the manual labeling process is expensive, time-consuming, and requires the expertise of the data. Therefore, the existing supervised learning algorithms are unable to take advantage of the abundance of real-world unlabeled data. Thus, relying on …