Engineering Commons^™

H-Nobs: Achieving Certified Fairness And Robustness In Distributed Learning On Heterogeneous Datasets, Guanqiang Zhou, Ping Xu, Yue Wang, Zhi Tian

Electrical and Computer Engineering Faculty Publications

Fairness and robustness are two important goals in the design of modern distributed learning systems. Despite a few prior works attempting to achieve both fairness and robustness, some key aspects of this direction remain underexplored. In this paper, we try to answer three largely unnoticed and unaddressed questions that are of paramount significance to this topic: (i) What makes jointly satisfying fairness and robustness difficult? (ii) Is it possible to establish theoretical guarantee for the dual property of fairness and robustness? (iii) How much does fairness have to sacrifice at the expense of robustness being incorporated into the system? To …

Predicting Motorcycle Crash Injury Severity In Utah: Utilizing Machine Learning Methods, Isaac Baah, Milan Zlatkovic, Mohamed M. Ahmed

Kentucky Transportation Center Presentations

No abstract provided.

Accurate Knowledge Of Roadway Horizontal And Vertical Alignment Is Essential, Bekir Bartin, Mojibulrahman Jami, Kaan Ozbay

Kentucky Transportation Center Presentations

No abstract provided.

Maximizing Mining Operations: Unlocking The Crucial Role Of Intelligent Fleet Management Systems In Surface Mining’S Value Chain, Arman Hazrathosseini, Ali Moradi Afrapoli

Earth and Environmental Sciences Faculty Publications

On the one side, the operational expenses of mining enterprises are showing an upward trend; and on the other side, conventional mining fleet management systems (FMSs) are falling short in addressing the high-dimensionality, stochasticity, and autonomy needed in increasingly complex operations. These major drivers for change have convinced researchers to search for alternatives including artificial-intelligence-enabled algorithms recommended by Mining 4.0. The present study endeavors to scrutinize this transition from a business management point of view. In other words, a literature review is carried out to gain insight into the evolutionary trajectory of mining FMSs and the need for intelligent algorithms. …

Promoting Safe Use Of Amrs By Assessing Their Residual Risks And Safety-Related Functions, Tongzhou Zhang, Sabrina Jocelyn, Chun Hong Law, Ali Aidibe, Soumaya Yacout

Articles dans des actes de congrès

With Autonomous Mobile Robots (AMRs), the usual risks are those of collision between a worker and the AMR, or a body part trapped between the AMR and a fixed or moving obstacle. To bypass obstacles (human or otherwise), AMRs feature autonomous control systems activating safety-related functions based on machine learning or optimization algorithms. This paper proposes a roadmap to guide integrators and users on the safe implementation of two AMRs. To do so, a risk assessment of two identical AMRs was performed, using ISO 12100, in a test bed consisting of an Intelligent Cyber-Physical Systems laboratory warehouse at Polytechnique Montreal. …

Engineering Self-Efficacy And Spatial Visualization: Connecting The Spatial Dots, Katrina L. Carlson

Dissertations, Master's Theses and Master's Reports

Questions exist as to why students in the ENG 1002: Introduction to Spatial Visualization (Spat Vis) course, an intervention course at Michigan Technological University (MTU), have historically attained higher average grades in their first year STEM courses, such as Engineering I and II, calculus I and II, computer science, and chemistry courses. Research shows the retention rate, especially of women, is higher for students who have taken Spatial Visualization. One possible explanation for these observed benefits may be related to the students’ confidence in their ability (self-efficacy) to gain the engineering graphics skills needed to become an engineer. No work …

Algorithms For Coordinating Multiple Autonomous Vehicles Under Various Constraints With Emphasis On Workload Balancing, Abhishek Patil

Dissertations, Master's Theses and Master's Reports

This dissertation focuses on developing algorithms to solve the problem of coordinating multiple autonomous vehicles under various constraints, aiming to produce practical solutions for real-world applications. Built upon three journal publications addressing two coordination-related problems in different domains, this research document tackles the challenges of heterogeneity constraints and cable entanglement issues encountered by autonomous vehicle systems.

The first problem tackles task allocation and path planning for heterogeneous ground mobile vehicles operating in a 2D environment with asymmetric travel costs. By enhancing previous Primal-Dual approximation heuristic methods, novel techniques are introduced to manipulate dual variables and achieve balanced workload distribution, ultimately …

Analysis Of Factors Affecting Engine Knock In Modern Spark Ignition Gasoline Engines To Improve Engine Efficiency, Siddharth Gopujkar

Dissertations, Master's Theses and Master's Reports

Spark ignition engine knock is an issue that has persisted for over a century. Knock prevents the spark in an SI engine to be deployed at the most optimum time in the engine cycle, which leads to a decrease in the efficiency of the engine. A reduction in the efficiency directly translates to increased emissions and fuel costs. This research looks at experimental works to analyze the impact of three separate factors on knock in an effort to minimize it and improve engine efficiency.

The first factor is fuel chemistry – how Research Octane Number (RON) and Motor Octane Number …

Solidification Cracking In Additively Manufactured High Entropy Alloy Nicrcofe With V Or W, Andrew R. Walch

Dissertations, Master's Theses and Master's Reports

Recent sustainability efforts have reduced CO2 emissions from electricity generation by increasing operating temperatures and pressures in advanced ultra-supercritical (A-USC) power plants. However, the extreme operating conditions in these A-USC power plants require new high temperature materials. The recent study of nickel-based high-entropy alloys (HEA) represents the latest approach to meet the specifications for A-USC operations. A CALculation PHase Diagram (CALPHAD) driven Bayesian optimization method was used to design the composition of alloys composed of NiCrCoFe with V or W, which were fabricated via wire-based additive manufacturing to enable high throughput development. Although the alloy was designed to minimize solidification …

Specification And Implementation Of Arm Isa Using Adl, Jonathan A. Rabideau

Dissertations, Master's Theses and Master's Reports

In this project, we implement a specification of ARM ISA using the ADL system. Current progress includes 79% of planned instructions, 79% of which are complete for our purposes, and also includes components such as a proxy kernel and a temporary preprocessor. Simple programs compiled directly from gcc can already be properly assembled and simulated using the tools generated by ADL. Once the specification is complete, we can integrate it with other components for testing ideas and theories. This report covers our progress and future plans regarding this project, including some background material and problem solutions.

Enhancing Iot Security: Optimizing Anomaly Detection Through Machine Learning, Maria Balega, Waleed Farag, Xin-Wen Wu, Soundarararjan Ezekiel, Zaryn Good

Computer Science Articles

As the Internet of Things (IoT) continues to evolve, securing IoT networks and devices remains a continuing challenge. Anomaly detection is a crucial procedure in protecting the IoT. A promising way to perform anomaly detection in the IoT is through the use of machine learning (ML) algorithms. There is a lack of studies in the literature identifying optimal (with regard to both effectiveness and efficiency) anomaly detection models for the IoT. To fill the gap, this work thoroughly investigated the effectiveness and efficiency of IoT anomaly detection enabled by several representative machine learning models, namely Extreme Gradient Boosting (XGBoost), Support …

A Robust Form Understanding System Using Graph-Based Neural Network, Chavin Chuangchaichatchavarn

Chulalongkorn University Theses and Dissertations (Chula ETD)

In this work, we address the challenge of form understanding in real-world documents affected by OCR noise and layout uncertainty. We introduce TONDFU, a bilingual Thai and English dataset consisting of official documents such as vehicle registrations and utility bills, annotated for entity labelling and entity linking. We also introduce a noisy character feature extractor that captures lexical and spatial patterns to improve the model's robustness against noisy textual content. This feature is integrated with geometric, visual, and semantic features in the graph-based model. Experiments show that the noisy character feature outperforms the frequency histogram baseline, and with pretraining on …

Muscle Activation And Coordinated Movements Of Infant Rolling, Danielle N. Siegel, Safeer F. Siddicky, Wyatt D. Davis, Erin M. Mannen

Mechanical and Biomedical Engineering Faculty Publications and Presentations

Rolling is a critical step of infant development, encouraging muscle coordination and enabling independent exploration. Understanding muscle activity during infant rolling movements on a flat surface is necessary to more fully characterize how the rolling milestone is achieved. The purpose of this study was to determine infants’ muscle activation throughout roll initiation for six previously established coordinated movements. Thirty-eight healthy infants (age: 6.5 ± 0.7 months; 23M/15F) were enrolled in this IRB-approved in-vivo biomechanics study. Surface electromyography sensors recorded muscle utilization from the erector spinae, abdominal muscles, quadriceps, and hamstrings while infants rolled. Each rolling movement was categorized as one …

Performance Reliability And Safety Index For Region Vii Highways, Aemal J. Khattak, Mm Shakiul Haque, Li Zhao

Mid-America Transportation Center: Final Reports and Technical Briefs

Roadway performance measures are used for operating, planning, and design purposes to improve services for road users. Simple statistics of the performance metrics such as measures of central tendency (e.g., mean, median) are often used for making operational and planning decisions. Even though such statistics are useful, they are, more than often, not comprehensive enough for making insightful operational and planning decisions. Consequently, performance reliability measures, which attempt to capture the day-to-day variability in performance, have the potential to benefit the areas of operation, planning, and design. Traffic data from Intelligent Transportation System facilitates the measurement of the reliability of …

Inconclusive Objects: Sample Making As Inquiry-Driven Practice, Etta Sandry

Textile Society of America: Symposium Proceedings

In Inconclusive Objects: Sample Making as Inquiry-driven Practice I will discuss craft sample making as a valuable practice of critical thought and experimentation. Craftspeople, artists, and designers make samples to practice technical skills and to demonstrate results on small scales. It is process work, typically used to learn a new technique or a hidden step on the way to a final product. However, as a cumulative process of research, material probing, and play, sampling is a critical aspect of design and creation. What can a closer study of sampling reveal about creative practice and knowledge production in craft and beyond? …

Economic, Environmental, And Social Characteristics Of E-Scooters On Lsu Campus, Mostafa Elseifi, Md Afif Rahman Chowdhury

Data

This study provides a comprehensive evaluation of e-scooter technology, with a focus on its environmental, economic, and social impacts, particularly within university settings at the Louisiana State University (LSU) campus. A life-cycle assessment (LCA) was conducted to assess the environmental sustainability of e-scooters, encompassing all stages from raw materials acquisition to end-of-life. The analysis determined that the total CO₂ emissions for an e-scooter over a five-year lifespan were approximately 389 kg, with significant reductions possible by improving the lifespan of e-scooters, integrating recycled materials, and utilizing renewable energy sources for charging batteries. The study also compared the environmental and …

Influence Of Nanofiller Surface Treatment On Mechanical Properties Of Pyrolyzed Ceramic Nanocomposites, Laxmi Sai Viswanadha, Chenglin Wu, Jeremy Watts, Mohammad Naraghi

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

The brittleness and limited plastic deformation of ceramics restrict their applications. In this work, the effects of pristine (CNTP), functionalized (CNTOH, CNTCOOH) and silanized MWCNTs on the mechanical properties of Silicon Carbide obtained by pyrolyzing polycarbosilane SMP-10 were studied. Functionalization with 3-glycidoxypropyltrimethoxy was analyzed through dispersion stability, zeta potential, and EDS analyses, revealing increased silicon content and colloidal stability in silanized MWCNTs. However, silanized MWCNTs led to reduced mechanical properties in composites, while untreated MWCNTs (CNTP, CNTOH, CNTCOOH), showed a substantial increase in modulus by 74.2 %, 86.9 %, and 30.5 %, respectively. The observed enhancement in mechanical properties exceeded …

Approximating Discrimination Within Models When Faced With Several Non-Binary Sensitive Attributes, Yijun Bian, Yujie Luo, Ping Xu

Electrical and Computer Engineering Faculty Publications

Discrimination mitigation with machine learning (ML) models could be complicated because multiple factors may interweave with each other including hierarchically and historically. Yet few existing fairness measures are able to capture the discrimination level within ML models in the face of multiple sensitive attributes. To bridge this gap, we propose a fairness measure based on distances between sets from a manifold perspective, named as ‘harmonic fairness measure via manifolds (HFM)’ with two optional versions, which can deal with a fine-grained discrimination evaluation for several sensitive attributes of multiple values. To accelerate the computation of distances of sets, we further propose …

Dynamics Of Cell Membrane Lesions And Adaptive Conductance Under The Electrical Stress, Mantas Silkunas, Olga N. Pakhomova, Giedre Silkuniene, Andrei G. Pakhomov

Bioelectrics Publications

Exceeding physiological limits of the cell membrane potential compromises structural integrity, enabling the passage of normally impermeant solutes and disrupting cell function. Electropermeabilization has been studied extensively at the cellular scale, but not at the individual membrane lesion level. We employed fast total internal reflection fluorescence (TIRF) imaging of Ca(2+) entry transients to discern individual lesions in a hyperpolarized cell membrane and characterize their focality, thresholds, electrical conductance, and the lifecycle. A diffuse and momentary membrane permeabilization without a distinct pore formation was observed already at a -100 mV threshold. Polarizing down to -200 mV created focal pores with a …

Studying Ht-Mcss In Ieee 802.11n Networks Via Simulations, Jun Peng

Electrical and Computer Engineering Faculty Publications

We study the High Throughput Modulation and Coding Schemes (HT-MCS) in IEEE 802.11n In networks in this paper. The HT-MCSs are designed to deal with various channel conditions in the networks. We used ns-3 simulations to study HT-MCS 0 to HT-MCS 31 under various radio propagation models. The simulation results are presented in this paper. In our simulations the HT-MCSs were tested in networks with Friis, Nakagami, and log-distance propagation models for studying their performance under various signal attenuation and fading effects. The HT-MCSs provided consistent performance under most channel conditions in our simulations. However, they showed significantly degraded performance …

Integrated Building Assessment To Enable Retrofit Design, Fabrication And Verification: A Drone-Based Approach, Paul Crovella

SyracuseCoE

Buildings account for approximately 40% of annual greenhouse gas emissions in the United States. In New York State, approximately 80% of residential and commercial buildings were constructed before the emerging energy codes in the 1970s [1]. Therefore, there is a significant need for affordable and scalable building retrofit solutions with low embodied emissions to meet the targets in the NYS Climate Leadership and Community Protection Act of 2019. A building energy audit is the initial step to quantify a building's energy performance and reveal its energy-saving potential. However, performing a high-level energy audit requires detailed field analysis and is often …

Development Of A High-Pressure Infiltration Process For Phenol–Formaldehyde Matrix Composites, Samuel Weiler, Patrick Schwartzkopf, Henry Haffner, K. Chandrashekhara

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Phenol–formaldehyde (phenolic) thermosets are known for excellent heat and chemical resistance, high flame retardance, and good mechanical performance. However, phenolics are also known for their high brittleness, and tendency to form voids, due to a condensation reaction forming water during curing. These voids can decrease the mechanical performance of the resultant phenolic composite and introduce undesirable performance characteristics. This work aims to develop a technique that uses high-pressure infiltration to obtain dense phenolic matrix composites, with commercially available resin and fiber reinforcement. The high-pressure system developed in this work is compared to a conventional low-pressure resin infusion technique, and the …

Development Of Digital Twin For Fdm Printer With Preventive Cyber-Attack And Control Algorithms, Md Hazrat Ali, Asad Malik, Nursultan Jyeniskhan, Muhammad Arif Mahmood, Essam Shehab, Frank Liou

Mechanical and Aerospace Engineering Faculty Research & Creative Works

This paper presents a developed model of a Digital Twin (DT) for a fused deposition modeling (FDM) printer, real-time defect detection, and proposed frameworks for preventing cyber-attacks in real-time. It also highlights a model predictive control (MPC) algorithm for controlling the material feed based on a real-time feedback system. The system is designed and developed based on DT, and MPC with integrated machine learning (ML) algorithms to establish real-time process control and enhance the safety and reliability of the physical plant. ML algorithm is used for anomaly detection based on the convolutional neural network (CNN) model. The developed system can …

Hyper-Thick Electrodes For Lithium-Ion Batteries Enabled By Micro-Electric-Field Process, Tazdik Patwary Plateau, Gracie Boyer, Jonghyun Park

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Increasing electrode thickness is a key strategy to boost energy density in lithium-ion batteries (LIBs), which is essential for electric vehicles and energy storage applications. However, thick electrodes face significant challenges, including poor ion transport, long diffusion paths, and mechanical instability, all of which degrade battery performance. to overcome these barriers, a novel micro-electric-field (μ-EF) process is introduced that enhances particle alignment during fabrication with reduced distance between anode and cathode. This process produces hyper-thick (≈700 µm) electrodes with low tortuosity and improved ion diffusion. the μ-EF electrodes achieve high areal capacities (≈8 mAh cm−2), while maintaining power density and …

Kinetic Simulations Of Dust Grain Charging Near Surfaces Under Various Plasma Conditions, David Lund, Daoru Han

Mechanical and Aerospace Engineering Faculty Research & Creative Works

This paper presents fully kinetic numerical investigations of the charging of spherical and irregular shaped dust grains in multiple configurations utilizing the Parallel Immersed-Finite-Element Particle-in-Cell Dusty (PIFE-PIC-D) framework. The simulations account for both surface charging and in-depth charging of the dust grains immersed in various ambient plasma conditions near a floating or biased surface. PIFE-PIC-D explicitly resolves the geometrical and material properties (permittivity and photoelectron emissions) of each individual dust grain. PIFE-PIC-D will model self-consistently the unsteady and stochastic charging of grains directly from the deposition of electrons and ions onto them. PIFE-PIC-D will also simulate the charging of irregular-shaped …

Development And Characterization Of An Ion Source To Simulate Low Earth Orbit Plasma Environment In A Vacuum Chamber, Emmanuel Kofi Asuako Wie-Addo, Scott Lucas, Daoru Han

Mechanical and Aerospace Engineering Faculty Research & Creative Works

A magnetic filter-type plasma source has been installed and characterized in a 10 ft long x 6ft diameter vacuum chamber. The plasma source was characterized with a fleet of diagnostic probes including an in-house cylindrical Langmuir probe. This resulted in the measurement of key plasma parameters such as electron number density, electron temperature, plasma potential, Debye length, ion current density, streaming ion energy, ion number density, and streaming ion velocity. By varying the flow rates, discharge current, and distance, single points measurements were taken at three locations downstream of the plasma source to establish an optimum operating condition to simulate …

Investigating Bistable Dynamics Of Coupled Oscillators With Similarities To Neural Activity In Epilepsy, Nilay Kant, Ranjan Mukherjee

Mechanical and Aerospace Engineering Faculty Research & Creative Works

The dynamics of an oscillator, which exhibits a stable equilibrium and a stable limit cycle, is investigated. We refer to it as a bistable oscillator unit (BOU) and show that two coupled BOUs (CBOUs) exhibit dynamics analogous to neural activity patterns in epilepsy, including healthy, localized, and fully spread epileptic states. By treating each CBOU as an independent system influenced by the state of the other, we establish local input-to-state stability near the equilibrium and the limit cycle and estimate the ultimate bounds of the trajectories. Our analysis identifies the domain of the initial conditions and estimates the coupling parameter …

Development And Evaluation Of Blind Spot Detection Safety System Based On 2d-Lidar Technology As An Optimization For Adas Systems, Timothy J. Sutton

Honors College Theses

Distracted driving poses a significant safety hazard and will only exacerbate as the number of modern-day distractions increases. To mitigate this problem, Advanced Driver Assistance Systems (ADAS) features, such as blind spot detection, have been pivotal for the safer operation of vehicles. Towards the same objective, the goal of this research is to utilize 2D LiDAR sensors to create a blind spot detection system that will detect objects and surfaces that are outside of the driver’s field of view. A comparative analysis was conducted by developing a 2D LiDAR-based system utilizing NVIDIA Jetson Orion Nano and Python alongside an ultrasonic-based …

The Hazard Prediction Problem, Mary E. Helander, Brendan Smith, Sylvia Charchut, Erika Swiatowy, Calvin Nau, Gregory Cavaretta, Timothy Schuler, Adam Schunk, Héctor Ortiz-Peña

Social Science - All Scholarship

This work formulates the hazard prediction problem while addressing the research question: Can machine learning create a model to automatically recognize patterns that correspond to hazard state conditions during a mission-critical operation? Supervised learning models were trained and tested on data observed from mission simulators, which allowed for safe observation of dynamic system states and undesirable casualty events. The prediction task was formulated as a binary classification problem, producing the probability of being in a hazard state at time t and providing situational awareness of a possible imminent loss. Several modeling architectures were investigated: neural networks, logistic regression, a support …

Optimizing Anaerobic Digestion Of Food Waste And Industrial Sludge For Biogas Production, Emma R. Murray

Honors College Theses

Food waste is a major contributor to municipal solid waste, and its disposal at landfills has significant environmental and economic impacts. Opportunities need to be identified for the sustainable management of food waste at large scales. Pulp and paper mills throughout the country commonly use anaerobic digestion to treat their waste sludge; however, this sludge can consist of over 50% lignin, which is slowly or negligibly digested. This research aims to explore the sustainability of co-digestion of food waste and paper mill sludge. Lab-scale studies were used to assess the feasibility of anaerobic digestion by measuring treatment efficiencies and biogas …