Engineering Commons^™

Developing Imitation Learning Policies By Exploring Diffusion Framework For Quadcopters, Likith Swamireddy

Theses and Dissertations

Imitation learning offers a promising approach in developing near-optimal onboard policies for nonlinear systems, by learning from computationally expensive trajectory planners during offline training. Existing State-of-the-art methods commonly learn deterministic mappings from observations to actions by training fast neural networks to mimic the expert demonstrations, which can limit their ability to capture the action distribution, when the policy is trained on multiple possible expert commands to reach the same target, and this may lead to unsafe scenarios in the obstacle environments. To address these challenges, this study explores Diffusion Models (DMs) as a feedback controller for quadcopters that learns the …

Enhancing Agricultural Sustainability Under Climate Change: A Multi-Scale Framework Integrating Climate Extremes, Resource Efficiency, And Data-Driven Modeling, Shahryar Fazli

Computational and Data Sciences (PhD) Dissertations

Agricultural systems are increasingly challenged by climate variability, where shifting temperature regimes, hydrological variability, and the rising frequency of compound and cascading extremes threaten global food security and resource sustainability. Addressing these challenges requires integrated frameworks that bridge biophysical monitoring, predictive modeling, and adaptive decision-making. This dissertation develops a data-driven, multi-scale framework to quantify and enhance agricultural resilience by integrating remote sensing, climate analytics, and machine learning across the United States, with a focus on California and the Western U.S.

First, hyperspectral and thermal remote sensing data from EMIT and OpenET are integrated to characterize crop nitrogen–water interactions and assess …

Evolving Ai Integration In Complex Medical Decision-Making And Multidisciplinary Transplant Care: A Systematic Review Of Human-Ai Collaboration, Rachel L. Dzieran, Cihan H. Dagli, Robert J. Marley

Engineering Management and Systems Engineering Faculty Research & Creative Works

Purpose of Review: Artificial intelligence (AI) in healthcare has evolved dramatically from early expert systems, which were initially considered replacements for clinical judgment, to today's collaborative frameworks that aim to augment physician decision-making. This evolution is particularly crucial in domains such as transplant surgery, where decisions carry irreversible consequences and require the integration of complex, often ambiguous data. Drawing on peer-reviewed literature from 2019 to 2025, we conducted a systematic review that analyzed key elements distinguishing successful human-AI partnerships from those that fail. Recent Findings: The ideal balance incorporates human expertise into AI systems through weighted integration approaches, rather than …

Robust Hybrid Tree-Based Machine Learning-Assisted Optimization Of Well Parameters To Reduce Asphaltene Precipitation Risk In Oil Fields, Malek Jalilian, Madani, Alireza Keshavarz, Stefan Iglauer, Abbas Khaksar Manshad, Amir H. Mohammadi

Research outputs 2022 to 2026

Asphaltene precipitation is a persistent flow-assurance issue in carbonate oil wells, leading to increased intervention frequency and production losses. This study utilizes multi-decade surveillance and operational data from a mature carbonate field (1983–2023) to train and optimize five tree-based models: Decision Tree, Random Forest, Extra Trees, Gradient-Boosting Decision Tree, and CatBoost, employing a Tree-Structured Parzen Estimator. These models, constrained by operational parameters, were integrated to identify optimal settings that can minimize the frequency of asphaltene precipitation and the subsequent cleanups. The novelty of this research lies in the direct integration of interpretable tree ensembles with an optimizer that adheres to …

Nanofluid-Assisted Enhanced Sealing Containment Of Caprocks For Efficient Geological Co2 Storage, Muhammad Ali, Nurudeen Yekeen, Mujahid Ali, Nilanjan Pal, Mohammed Al Kobaisi, Alireza Keshavarz, Hussein Hoteit

Research outputs 2022 to 2026

Millions of tons of CO2 are stored annually in geological formations in order to reduce greenhouse gas emissions, relying on caprock as a seal to prevent CO2 leakage. The wettability of caprock is crucial for its effectiveness, and can be altered by organic acids present in the storage media. The present study investigates the impact of stearic acid on the CO2 wettability of shale, along with the potential of alumina nanofluids to reverse this effect. Using contact angle measurements, X-ray diffraction, and other analytical methods, the study reveals that stearic acid increases the CO2 wettability of shale, making it more …

A Consolidated Mössbauer Spectrum For Undisturbed Soils From Gusev Crater, Mars, Paulo De Souza

Research outputs 2022 to 2026

Mossbauer analyses from MIMOS II data collected on Mars provided breakthrough insights into iron mineralogy of the planet’s surface. Resulting from limited measurement time, science operations were divided among multiple instruments, restricting the detection of minor Fe-bearing mineral phases (below 5% WSA). MIMOS II records Mössbauer spectra at 10 K intervals, and to enhance signal to noise ratios, spectra from different intervals were combined, when applicable. For instance, temperature variations between 190 K and 270 K in basaltic samples do not significantly impact the spectra, allowing data from all intervals to be integrated. By combining MIMOS II data from various …

Characterization Of Droplet Size And Their Distribution Using Low Operating Conditions In A Rotary Bell Cup Spray-Field Via Shadowgraph, Murad Ali Channa, Mehdi Khiadani, Yasir M. Al-Abdeli

Research outputs 2022 to 2026

This study characterises droplet size and distribution across axial (x/D) and radial (r/D) locations in the spray field of a rotary bell-cup atomizer under low operating conditions. Shadowgraph was used to measure droplet size and distribution at 16 dimensionless locations in both directions. Two low rotational-speed ranges, 3–3.5 krpm (case I) and 8–8.5 krpm (case II), were achieved using shaping-air of 30–100 L/min under turning-air of 50 and 60 L/min, respectively, at a constant flow rate of 0.5 L/min. The Sauter mean droplet diameter ranged from ∼ 45–98 µm in case I and narrowed to ∼ 45–85 µm in case …

Dataset For Integrity Attacks On Time Synchronized Synchrophasor Data, Taylah Griffiths, Mohiuddin Ahmed, Chadni Islam

Research outputs 2022 to 2026

Phasor measurement units, also known as synchrophasors, are a vital component within smart grids to determine the stability of the grid. These devices send synchrophasor data to phasor data concentrators that collate and analyse the data. Recently, synchrophasor communication data has become beneficial for the research community. However, datasets covering cyberattacks on synchrophasor data are not public. Having access to this data would aid in investigating mitigations against cyberattacks. This paper describes a public specialized dataset, known as ECU-PMU-FDI/TSA. The dataset contains synchrophasor communication data for cybersecurity mitigation testing. Three hours of communication data was captured, from a simulated testbed. …

Thermophysical Characteristics Evaluation Of Carbon And Metal Oxide-Based Hybrid Phase Change Materials For Thermal Energy Storage Applications, Utpol K. Paul, Md Shahriar Mohtasim, Md Golam Kibria, Barun K. Das

Research outputs 2022 to 2026

Phase change materials (PCM) have limited practical applications due to their low heat conductivity. The main goal of this work is to improve the thermophysical properties of PCM by adding different compositions of hybrid carbon and metal oxide-based nanoparticles for incorporation into hybrid photovoltaic/thermal solar systems to accommodate the built environment and thermal energy storage as a building envelope for thermal comfort. Three types of nanoparticles have been selected including Graphene, Al2O3, and ZnO and synthesized samples with the combinations of 0.5 wt% Al2O3 + 2 wt% ZnO, 0.75 wt% Al2O3 + 2 wt% ZnO, 1 wt% Al2O3 + 2 …

Green Waste Biochar And Plant Growth-Promoting Bacteria Enhance Tomato Growth Under Combined Nutrient Deficiency And Salinity Stress, Soumaya Tounsi-Hammami, Munawwar Ali Khan, Mahra Alqemzi, Salama Ali Almehairi, Aneesa Rasheed Anwar

All Works

This study characterized a green-waste-derived biochar from date palms and ghaf trees and investigated its potential as a soil amendment with halotolerant Bacillus spp. to improve tomato seedling quality under dual stress of salinity and nutrient deficiency. Biochar was produced through pyrolysis at 450 °C and then characterized for yield, pH, electrical conductivity, proximate analysis, surface morphology, energy-dispersive X-ray spectroscopy, and heavy-metal content. Its effectiveness was tested both alone and in combination with a Bacillus sp. mix, using a completely randomized design with varying NPK fertilizer levels and saline irrigation. Tomato seedlings were evaluated 45 days after planting for various …

Flight System Development For A Bio-Inspired Rotating Empennage Experimental Unmanned Aerial Vehicle, Ashton Gilbert

All Graduate Theses and Dissertations, Fall 2023 to Present

Unmanned aerial systems, commonly known as drones, are increasingly used for research, environmental monitoring, and testing new aircraft concepts. Many commercially available drone flight controllers work well for standard aircraft designs, but they can be difficult to adapt for experimental vehicles that use unconventional control methods or require detailed monitoring during testing. This creates challenges for researchers who need flexible and reliable tools to safely evaluate new flight technologies. This work presents the development of a custom drone flight control system designed specifically for experimental research. The system integrates custom electronics and software that allow researchers to easily modify control …

Scalable Quality Assessment Of Ground Motion Records Via Interpretable Deep Learning Architectures, Ali Montazeri Namin

All Graduate Theses and Dissertations, Fall 2023 to Present

Earthquake engineers rely on accurate recordings of ground shaking to design safe and resilient buildings. However, sorting through thousands of these recordings to find the reliable ones and throwing out those ruined by sensor errors or background noise is traditionally done by hand. Because modern seismic networks collect massive amounts of earthquake data every day, this manual checking process is much too slow. To fix this, researchers are turning to artificial intelligence to automatically check the quality of these recordings.

While artificial intelligence offers a fast solution, there are limitations. These computer models can become massive and expensive to run, …

Cascaded Pid Control Of A Passive Bio-Inspired Rotating Empennage Aircraft, Zachary T. Jenkins

All Graduate Theses and Dissertations, Fall 2023 to Present

This thesis presents an autopilot for a Bio-Inspired Rotating Empennage (BIRE) aircraft. Most airplanes control yaw using a vertical stabilizer and a moving rudder, a design that has stayed essentially unchanged for nearly a century but adds weight, drag, and surface area to the aircraft. The BIRE removes the vertical tail entirely and instead allows the horizontal tail to rotate about the body x-axis of the airplane; tilting the horizontal tail produces a yaw force, doing the job of a rudder. This thesis first develops a controller for the BIRE where the empennage angle is commanded directly, then extends …

Physics-Data-Hybrid Modeling And Control For Vtol Aircraft, Samantha Dawn Burton

All Graduate Theses and Dissertations, Fall 2023 to Present

Vertical takeoff and landing (VTOL) aircraft are hybrid aircraft having both a helicopter flight mode and a fixed-wing plane flight mode. The objective of this dissertation is to examine and evaluate a new approach, physics-data-hybrid (PDH), for VTOL aircraft modeling and control. The PDH approach combines traditional, physics-based methods (Newtonian mechanics) with data driven methods (a machine learning technique called a Koopman operator model) to create an efficient and accurate modeling and control framework—mitigating the disadvantages of the purely physics-based and purely data-driven methods. The PDH approach is used to model scenarios when the VTOL aircraft transitions between helicopter mode …

Efficient Charging Scheduling Through Coordination Of Electric Vehicle Platoons And Charging Stations, Liwan Qi, Bochun Wu, Shoubo Li, Yi Gong, Wei Ni

Research outputs 2022 to 2026

This paper focuses on charging allocation in a vehicle-to-infrastructure (V2I) communications-enabled electric vehicle (EV) network with heterogeneous traffic flows, where manned EVs and EV platoons coexist, and each EV platoon may have a different size and travel speed. In such a network, hybrid traffic flows pose significant challenges since platoons with multiple EVs can easily cause severe station overloading and increase the total time cost for charging service, particularly when large platoons occur. To tackle this issue, a centralized approach is proposed to plan charging allocation and optimize the velocities of manned EVs and EV platoons with the assistance of …

Natural Gas Hydrate Production And Co2 Storage Via Clathrate Hydrates: Challenges And Opportunities, Abdirahman Hassan Mohamed, Aliyu Adebayo Sulaimon, Haylay Tsegab, Bhajan Lal, Stefan Iglauer, Muhammad Ali

Research outputs 2022 to 2026

One of the most demanding environmental and economic challenges of this era is that of supplying the increasing global energy demand while reducing and neutralizing the CO2 footprint. In this regard, the provision of a secure and diversified energy supply and the sequestration of CO2 into geological formations are currently of great significance. However, the buoyant nature of CO2 under the temperature and pressure conditions of typical geological sites leads to the risk of CO2 leakage, and thus necessitates long-term monitoring. Therefore, the hydrate-based sequestration of CO2 beneath oceanic sediments has become a desirable alternative to conventional geologic sequestration in …

Thermally-Driven Equiaxation Of Columnar Grains And Α″ Nanoparticles Enabling Isotropic Β-Ti Alloys With Strength–Ductility Synergy Manufactured By Laser Powder Bed Fusion, Lei Qin, Shengfeng Zhou, Jianbo Jin, Huan Yang, Kunmao Li, Cheng Deng, Yujie Yuan, Lai-Chang Zhang

Research outputs 2022 to 2026

The transient thermal cycling characteristics during laser powder bed fusion (LPBF) induce elemental segregation and columnar growth, resulting in significant mechanical anisotropy and strength–ductility trade-off in titanium alloys. In order to break through this bottleneck, this work proposes a strategy of inducing equiaxation of columnar grains through recrystallization and regulating the morphology of the α″ phase (from an acicular shape to nanoparticles) to synergistically optimize the uniformity, strength and ductility of LPBF-fabricated Ti-35Nb-5Cu-xMo (Ti355x, x = 0, 1, 2, 4 wt%) alloys. The results demonstrate that LPBF-fabricated Ti355x alloys display equiaxed/columnar microstructures with significant mechanical anisotropy and strength–ductility trade-off. After …

Aluminium Based Hydrotalcite Like Compounds For Direct Lithium Extraction: A Critical Review On Crystallography And Structure Selectivity Design, Amin Sarmadi, Yasaman Boroumand, Mohammad Mahdi A. Shirazi, Magdalena Wajrak, Shayan Abrishami, Matthew Myers, Muhammad Rizwan Azhar, Amir Razmjou

Research outputs 2022 to 2026

The accelerating demand for lithium is intensifying interest in direct lithium extraction (DLE) technologies that minimise the environmental footprint of conventional solar evaporation ponds. Among sorbent-based DLE concepts, aluminium based hydrotalcite like compounds (Al-HTlc), particularly lithium aluminium layered double hydroxide (LiAl-LDH), are considered the most mature sorbent options, offering facile lithium (Li+) recovery and high selectivity in highly saline chloride based solutions. Both properties arise from a distinct intercalation mechanism wherein Li+ and its counter-anions (e.g. Cl− or SO42−) are co-intercalated into the crystal structure upon brine exposure. Li+ selectivity arises partly from a steric match between Li+ (ionic radius …

Scalable And Interpretable Fraud Detection In Fintech: Evidence From A Hybrid Tabnet–Graph Neural Network Framework, Haseeb Ahmad, Istiaque Bhuiyan, Tanvir Bhuiyan, Ariful Hoque

Research outputs 2022 to 2026

This study proposes a fraud detection framework for FinTech that integrates tabular deep learning with relational graph reasoning. We address a key limitation of state-of-the-art tabular fraud models: their assumption that transactions are independent, despite evidence that modern fraud often arises from coordinated infrastructure reuse (e.g., shared devices, email domains, accounts, and addresses). Using the IEEE-CIS Fraud Detection dataset (≈590,540 e-commerce transactions), we develop a hybrid architecture that combines TabNet, chosen for its built-in interpretability on tabular data, with a GraphSAGE encoder trained on a leakage-free heterogeneous entity graph constructed strictly from historical training transactions. Unseen entities in validation and …

Switching Configurations For Reconfigurable Battery Packs: A Review, Kenneth Tetteh Akiti, Quoc Viet Phung, Iftekhar Ahmad, Daryoush Habibi

Research outputs 2022 to 2026

Conventional Battery Energy Storage Systems (BESS) are designed with fixed cell topologies, where the electrical connection between cells remains unchanged throughout their life. Although Battery Management Systems (BMS) and conventional cell-balancing techniques are employed to monitor and maintain performance, they often fail to address key challenges such as uneven cell aging, capacity mismatch, and fault isolation that arise especially during the second life of a battery pack. These limitations can result in reduced efficiency, reduced reliability, and shortened battery life. As large-scale battery packs become increasingly common in Electric Vehicles (EVs), smart grids, robotics, and renewable energy systems, the limitations …