Computational Engineering Commons^™

Deep Learning For Multiple Unmanned Aerial Vehicle Coordination In Air Corridors, Liangkun Yu

Electrical and Computer Engineering ETDs

In the future, city skies will be filled with Unmanned Aerial Vehicles (UAVs) for rapid urban transport, including parcel deliveries and air taxis. NASA's Urban Air Mobility (UAM) envisions UAVs navigating air corridors. These virtual pathways ensure safety and compliance with regulations. However, current research on UAM practical applications is limited. This dissertation focuses on designing air corridors, developing UAV control systems, and ensuring the robustness of control algorithms against disturbances in real-world environments.

Our design features an air corridor system with horizontal lanes and on-off ramps, conceptualized as cylindrical spaces and tori, respectively. To enable each UAV to locally …

Development Of A Rule-Based Monitoring System For Autonomous Heavy Equipment Safety, Amirpooya Shirazi

Department of Construction Engineering and Management: Dissertations, Theses, and Student Research

Roadway construction work zones are constantly exposed to interactions among construction equipment, workers, and vehicles. Furthermore, ensuring safety in these areas is considered a challenging task due to the complexity of the environment. As shown in the rising trend of fatal accidents in roadway work zones, current OSHA regulations in construction safety are insufficient in effectively detecting unsafe situations and mitigating the risks. Furthermore, best practices, such as internal traffic control planning (ITCP), exhibit critical limitations requiring continuous monitoring of active work zones as well as adjustments to the site coordination plans due to the dynamic nature of work zone …

Safe And Efficient Operation Of Mobile Robots In Indoor Environments: A User-Centric Shared Control System With High-Level Navigation Capabilities, Ahmet Saglam

Electrical & Computer Engineering Theses & Dissertations

Hospitalization and isolation can be a traumatic experience for immunocompromised children, especially because they are separated from their families and friends. Social robots have been proposed as a way to improve the quality of care for children hospitalized in isolation by providing alternative means of social interaction and support. Remote control of such robots in a hospital setting, particularly where safety is a major concern, can be a daunting task for young patients.

This dissertation introduces a multilevel shared control system for mobile robots, specifically companion robots in hospital-like indoor spaces. The system integrates user inputs with algorithmic semi-autonomous control …

Auxiliary Diagnosis Of Dental Calculus Based On Deep Learning And Image Enhancement By Bitewing Radiographs, Tai Jung Lin, Yen Ting Lin, Yuan Jin Lin, Ai Yun Tseng, Chien Yu Lin, Li Ting Lo, Tsung Yi Chen, Shih Lun Chen, Chiung An Chen, Kuo Chen Li, Patricia Angela R. Abu

Ateneo Laboratory for Intelligent Visual Environments

In the field of dentistry, the presence of dental calculus is a commonly encountered issue. If not addressed promptly, it has the potential to lead to gum inflammation and eventual tooth loss. Bitewing (BW) images play a crucial role by providing a comprehensive visual representation of the tooth structure, allowing dentists to examine hard-to-reach areas with precision during clinical assessments. This visual aid significantly aids in the early detection of calculus, facilitating timely interventions and improving overall outcomes for patients. This study introduces a system designed for the detection of dental calculus in BW images, leveraging the power of YOLOv8 …

Problem Solving / Javascript Programming, Sarah Zelikovitz, Orit D. Gruber

Open Educational Resources

This Lab Experiment focuses on JavaScript Programming. Upon completing the lab, you will be able to understand the following:

· The definition of Algorithmic Problem Solving.

· The role of JavaScript in web pages.

· The concept of Iteration in computer programming.

Finding The Shortest Path Using Dijkstra’S Algorithm, Orit D. Gruber, Deborah Sturm

Open Educational Resources

This lab experiment explores an algorithm which is used to find the shortest path between two or more locations. After completing the lab, you will be able to answer the following questions in the final lab report:

1. What is an Algorithm?
2. What is a Graph ?
3. What is the purpose and operation of Dijkstra’s Algorithm ?

Aligning Language Models With The Human World, Ruibo Liu

Dartmouth College Ph.D Dissertations

The field of Natural Language Processing (NLP) has undergone a significant transformation with the emergence of large language models (LMs). These models have enabled the development of human-like conversational assistants (e.g., OpenAI's ChatGPT), and expert-level AI software engineering agents (e.g., Devin from Cognition Lab). However, these models face a fundamental challenge related to their training methodology. Predominantly trained on vast datasets scraped from the web, their self-supervised learning objective---predict missing tokens---unintentionally perpetuates the biases, inaccuracies, and sensitive information inherent in their training data. These issues lead to what is termed misaligned}behaviors and pose a significant hurdle in the development of …

Advancing Omnimodality: Expanding Human Creativity Through Adaptable And Accessible Multimodal Computing Systems, Joshua Urban Davis

Dartmouth College Ph.D Dissertations

Emerging technologies have given us a whole host of new ways for people to be

creative. From the immersive worlds of AR/VR to the synthesis powers of largelanguage

models and generative AI, these new tools hold the potential to reshape

human expression and creativity. But how can we ensure that these new modalities

are accessible to everyone, even those who aren’t able bodied? This thesis advocates

for a human-centered approach to the development of many-modal systems. I will

probe how our machines support, direct, or inhibit creativity as a mode of problem

solving through 6 novel multimodal prototype interfaces and …

The Effect Of Watts-Strogatz And Barabási-Albert Graphs On Memory Formation, Ethan Irick Wolfe

Master's Theses

Understanding higher level cognitive processes is a central problem in neuroscience. The Neuroidal model provides a useful framework for posing these problems in a computer science context. There has been significant recent work trying to understand memory capacity in the Neuroidal model but this work was done assuming that the network of neurons was an Erdos-Renyi random graph. However the network of neurons in the brain has been shown to exhibit small-world properties, which are not present in Erdos-Renyi graphs. In this research we explore replacing Erdos-Renyi graphs with Watts-Strogatz and Barabasi-Albert graphs in order to more accurately model the …

Extreme Image Transformations Improve Latent Representations In Machines, Girik Malik, Ennio Mingolla

MODVIS Workshop

Shuffling pixels in an image helps machines to learn a more robust object representation. To probe the strategies used by humans and machines for object recognition, we introduce Extreme Image Transformations (EITs). Machines rely heavily on exploiting low-level features like color and texture, so their performance degrades on out-of-distribution and adversarial inputs. Humans depend on high-level features like shapes and contours, making them relatively robust to image distortions. EITs systematically shuffle the pixels in an image, parameterized by the size of grids, probability of shuffle and binary block movement, distorting the structure of objects at both local and global levels. …

Explaining The Staircase Gelb Illusion, Simultaneous Contrast, And Perceptual Fading Of Stabilized Images With A Neural Model Driven By Fixational Eye Movements, Michael E. Rudd

MODVIS Workshop

A neural model of lightness computation driven by fixational eye movements is described and used to simulate various lightness phenomenon, including the Staircase Gelb illusion and its variants, simultaneous contrast, the Chevreul illusion, and perceptual fading of stabilized images. The model provides a precise account of the lightness matches from several experiments, with an overall error of only 1.5%. In the model, spatial maps of transient ON and OFF cell activations—produced as the eyes traverse the visual scene—are sorted by eye movement direction in visual cortex. At a subsequent processing stage, the activations within these maps are summed across space …

Neuro-Symbolic Commonsense Reasoning With Resistance To Data Poisoning: A First-Order Logic And Sub-Symbolic Embeddings Framework, Bryce Shurts, King-Ip Lin

Computer Science and Engineering Theses and Dissertations

Commonsense reasoning has long presented a hurdle between conversational agents and their ability to naturally engage with humans in conversation, as the infinitely dimensional nature of a dialogue’s topics presents a significant reasoning challenge in the study of Natural Language Understanding (NLU). Such a system must conceivably be able to act as a generalizable system for evaluating and reasoning about commonsense statements, problems, and queries: in this way, the agent can attempt to quantify the reasonability of a given input. We attempt to address this through the integration of an explainable neuro-symbolic system that leverages Logical Tensor Networks (LTNs) and …

An Attention Lstm U-Net Model For Drosophila Melanogaster Heart Tube Segmentation, Xiangping Ouyang

McKelvey School of Engineering Graduate Student Theses & Dissertations

Machine learning is commonly used in biomedical image analysis, as it allows automated image segmentation and identification that minimizes the need for tedious human involvement. Drosophila melanogaster is often used as a cardiac disease model, where optical coherence microscopy (OCM) is used to image and analyze its beating dynamics. As OCM often generates a large volume of images, automated image segmentation is necessary to quantify the heart beating efficiently. Our most recent heart segmentation model, FlyNet 2.0+, is a fully convolutional LSTM U-Net model. However, the performance of the model diminishes in the presence of artifacts, such as image reflection …

Securing The Skies: Safety-Constrained Decentralized Multi-Uav Coordination With Deep Reinforcement Learning, Jean-Elie Pierre

Electrical and Computer Engineering ETDs

In the dynamic landscape of autonomous aerial systems, the integration of uncrewed aerial vehicles (UAVs) has sparked a paradigm shift, offering unprecedented opportunities and challenges in collaborative decision-making and navigation. This thesis explores the application of multi-agent reinforcement learning (MARL) for the planning and coordination of UAVs in complex environments.

The first part of this thesis provides an introduction to single-agent reinforcement learning and MARL. We provide examples of the use of MARL for countering uncrewed aerial systems (C-UAS). We formulate the Counter-UAS problem as a multiagent partially observable Markov decision process (MAPOMDP), and we propose Multi-AGent partial observable deep …

Presence Of Atheromatous Plaques And Theirs Effects On The Blood Flow, Belhocine Mostefa Bm, Amrani Hichem Ah, Fedaoui Kamel Dr, Mazouz Hammoudi Mh

Emirates Journal for Engineering Research

The paper utilizes a finite element method to study both the blood flow and atheromatous plaques. Specifically, the COMSOL finite element package is employed to achieve a fluid model. COMSOL is a powerful finite element tool commonly used in various research and industrial domains to study multiphysics problems. The focus of the investigation is on the geometric aspects of the atheromatous plaques. The study considers different forms and arrangements of stenosis, taking into account the irregularities formed by various shapes of the plaques and the resulting flow patterns. The key findings of the research suggest that the pressure and velocity …

Star-Based Reachability Analysis Of Binary Neural Networks On Continuous Input, Mykhailo Ivashchenko

School of Computing: Dissertations, Theses, and Student Research

Deep Neural Networks (DNNs) have become a popular instrument for solving various real-world problems. DNNs’ sophisticated structure allows them to learn complex representations and features. However, architecture specifics and floating-point number usage result in increased computational operations complexity. For this reason, a more lightweight type of neural networks is widely used when it comes to edge devices, such as microcomputers or microcontrollers – Binary Neural Networks (BNNs). Like other DNNs, BNNs are vulnerable to adversarial attacks; even a small perturbation to the input set may lead to an errant output. Unfortunately, only a few approaches have been proposed for verifying …

Development And Simulation Of A Damage Assessment And Recovery Method For Critical Database Systems, Anthony Pham

Computer Science and Computer Engineering Undergraduate Honors Theses

With how much the world relies on technology and the critical database infrastructure that supports it, the infrastructures require efficient methods to detect and resolve suspicious database transactions, whether malicious or not. This paper focuses on an algorithm that detects and resolves malicious transactions in a database. The process begins with identifying suspicious transactions based on common patterns. When a transaction is flagged, the algorithm segments groups of suspicious transactions in separate log files, separating them for easy access. Within these segments, any dependent transactions that use data affected by the suspicious transactions will be stored there. After the transaction …

Enhancing Cybersecurity In Wireless Sensor Networks: Machine Learning, Blockchain And Future Perspectives, Ilemona Solomon Atawodi

Dissertations

Security in the Industrial Internet of Things encounters various security issues but the main issues can be broken down into three core issues: Availability, Integrity, and Confidentiality. Security challenges generally tend to be caused by a failure of the system in one of these areas or cause a failure in one of these areas. Therefore researching scalable solutions to these security issues is prudent to explore methods that could be applied to large-scale industrial IIoT with tens to hundreds of devices as well as small-scale systems on a tiny factory floor comprising of just a few devices. In our research, …

Automated Brain Tumor Classifier With Deep Learning, Venkata Sai Krishna Chaitanya Kandula

Electronic Theses, Projects, and Dissertations

Brain Tumors are abnormal growth of cells within the brain that can be categorized as benign (non-cancerous) or malignant (cancerous). Accurate and timely classification of brain tumors is crucial for effective treatment planning and patient care. Medical imaging techniques like Magnetic Resonance Imaging (MRI) provide detailed visualizations of brain structures, aiding in diagnosis and tumor classification^[8].

In this project, we propose a brain tumor classifier applying deep learning methodologies to automatically classify brain tumor images without any manual intervention. The classifier uses deep learning architectures to extract and classify brain MRI images. Specifically, a Convolutional Neural Network (CNN) …

Examining The Impact Of Customer Rfp Characteristics On Award Compliance, Laasya Ravipati

Data Science Undergraduate Honors Theses

In the context of intermodal transportation, understanding the dynamics of award compliance holds significant importance for operational efficiency and strategic decision-making. Award compliance refers to the percentage of awarded freight volume that is realized, indicating the extent to which contractual agreements are fulfilled. This analysis delves into the intricate relationship between customer characteristics and award compliance, aiming to provide valuable insights into the variability and predictability of compliance rates. By analyzing Request for Pricing (RFP) data and primary awarded freight volumes, the study seeks to address the need for more accurate volume estimations, crucial for sales planning, revenue projections, and …

Numerical Simulation Of Laser Induced Elastic Waves In Response To Short And Ultrashort Laser Pulses., Alireza Zarei

All Dissertations

In an era of intensified market competition, the demand for cost-effective, high-quality, high-performance, and reliable products continues to rise. Meeting this demand necessitates the mass production of premium products through the integration of cutting-edge technologies and advanced materials while ensuring their integrity and safety. In this context, Nondestructive Testing (NDT) techniques emerge as indispensable tools for guaranteeing the integrity, reliability, and safety of products across diverse industries.

Various NDT techniques, including ultrasonic testing, computed tomography, thermography, and acoustic emissions, have long served as cornerstones for inspecting materials and structures. Among these, ultrasonic testing stands out as the most prevalent method, …

Developing General Purpose Apps To Automate Image Analysis Of Wave-Augmented-Varicose-Explosion Atomization And Other Multi-Phase Interfacial Flows, Ethan Newkirk

Senior Honors Theses

Atomization involves disrupting a flow of contiguous liquid into small droplets ranging from one submicron to several hundred microns (micrometers) in diameter through the processes of exerting sufficient forces that disrupt the retaining surface tensions of the liquid. Understanding this phenomenon requires high-speed imaging from physical models or rigorous multiphase computational fluid dynamics models. We produce a MATLAB application that utilizes various methods of image analysis to quickly analyze and store mathematical data from detailed image analyses. We present a user with numerous tools and capabilities that provide results that deviate from 1.8% to 8.9% of the original image sequence …

Context Dependent Training Data Selection For Automatic Target Detection., Tylman Michael

Electronic Theses and Dissertations

An Automatic Target Detection (ATD) algorithm is capable of identifying the location of targets of interest captured by Infra-Red imagery in vastly different contexts. ATD is often a precursor in a 2-stage methodology in order to ascertain the location and nature of a target in both military and civilian applications. In order to train an ATD algorithm, a large amount of data from varied sources is required. One drawback of this requirement is that some sources of data may harm the performance of the method in different contexts. This thesis explores utilizing an unsupervised method to identify a subset of …

Quantifying Hurricane Effects On Housing: Evaluating Damage, Loss, And Shelter Demands Using Historical And Simulated Storm Tracks, Adish Deep Shakya

All Theses

This research introduces an advanced framework which employs parametric wind field models for peak wind speeds, and building fragility curves, loss functions, and demographic data to estimate for estimating housing damage and loss. The uninhabitable units immediate displaced households, short-term and long-term shelter need households are determined. with a particular focus on those eligible for FEMA assistance. The framework's validity is reinforced by a high correlation in the analysis of recent hurricane events between estimated numbers of displaced households and actual FEMA aid recipients, where FEMA aids about 20-60% of the predicted long-term displaced households. A novel application of the …

Data Engineering: Building Software Efficiency In Medium To Large Organizations, Alessandro De La Torre

Whittier Scholars Program

The introduction of PoetHQ, a mobile application, offers an economical strategy for colleges, potentially ushering in significant cost savings. These savings could be redirected towards enhancing academic programs and services, enriching the educational landscape for students. PoetHQ aims to democratize access to crucial software, effectively removing financial barriers and facilitating a richer educational experience. By providing an efficient software solution that reduces organizational overhead while maximizing accessibility for students, the project highlights the essential role of equitable education and resource optimization within academic institutions.

Stiffness And Strength-Based Concurrent Shape And Fiber Path Optimization Of Continuous Fiber Composites, Zhelong He, Ali Tamijani

Publications

A computer-implemented method of optimizing a computer model including a shape and a fiber path for a continuous fiber composite can include initializing a fixed finite element mesh. The method can also include creating an updated version of the shape and the fiber path by iterating updates of the shape and the fiber path using the fixed finite element mesh. The method can also include initializing an adapted finite element mesh on the updated version of the shape and the fiber path. The method can also include creating an optimized version of the shape and the fiber path by iterating …

Computational Modeling And Analysis Of Facial Expressions And Gaze For Discovery Of Candidate Behavioral Biomarkers For Children And Young Adults With Autism Spectrum Disorder, Megan Anita Witherow

Electrical & Computer Engineering Theses & Dissertations

Facial expression production and perception in autism spectrum disorder (ASD) suggest the potential presence of behavioral biomarkers that may stratify individuals on the spectrum into prognostic or treatment subgroups. High-speed internet and the ease of technology have enabled remote, scalable, affordable, and timely access to medical care, such as measurements of ASDrelated behaviors in familiar environments to complement clinical observation. Machine and deep learning (DL)-based analysis of video tracking (VT) of expression production and eye tracking (ET) of expression perception may aid stratification biomarker discovery for children and young adults with ASD. However, there are open challenges in 1) facial …

Development Of On-The-Fly Quasi-Steady State Approximation For Chemical Kinetics In Cfd, Abhinav Balamurugan

Doctoral Dissertations and Master's Theses

This study analyzes the feasibility of On-The-Fly Quasi-Steady-State Approximation (OTF-QSSA) application for solving chemical kinetics within Computational Fluid Dynamics (CFD) simulations, aiming to reduce the computational demand of detailed mechanisms. An algorithm that dynamically identifies and designates Quasi-Steady-State (QSS) species at specific grid locations and instances during the simulation was developed. With this information, our method pseudo-delays the advancement of concentrations for these QSS species—effectively setting their rate of concentration change to zero for a set number iteration before updating using the detailed mechanism and thereby omitting the computationally intensive processes typically required for their calculation during those skipped iteration. …

Revolutionizing Feature Selection: A Breakthrough Approach For Enhanced Accuracy And Reduced Dimensions, With Implications For Early Medical Diagnostics, Shabia Shabir Khan, Majid Shafi Kawoosa, Bonny Bannerjee, Subhash C. Chauhan, Sheema Khan

Research Symposium

Background: The system's performance may be impacted by the high-dimensional feature dataset, attributed to redundant, non-informative, or irrelevant features, commonly referred to as noise. To mitigate inefficiency and suboptimal performance, our goal is to identify the optimal and minimal set of features capable of representing the entire dataset. Consequently, the Feature Selector (Fs) serves as an operator, transforming an m-dimensional feature set into an n-dimensional feature set. This process aims to generate a filtered dataset with reduced dimensions, enhancing the algorithm's efficiency.

Methods: This paper introduces an innovative feature selection approach utilizing a genetic algorithm with an ensemble crossover operation …

A 4-Node Shell Finite Element Based On Assumed Bending And Membrane Strains For Static Analysis Of Plates And Shells, Sifeddine Abderrahmani

Emirates Journal for Engineering Research

In this paper the development of a new rectangular flat shell element is proposed. This element is called SBRPK-SBRIE. This element is used in the numerical analysis of thin structures based on the strain approach with linear elastic behavior. Combining bending and membrane elements yields the proposed element.The strain-based rectangular finite element for the thin plate bending element denoted SBRPK, and the strain-based membrane element denoted SBRIE. Several numerical examples have been conducted to assess the accuracy and reliability of the developed element compared with the theoretical results and other finite elements. Obtained results show its good performance compared to …