Engineering Undergraduate Research Theses and Honors Research Theses

Permanent URI for this collection

Undergraduate Research Theses and Honors Research Theses from the College of Engineering. More about the College of Engineering Honors Program can be found at:

Instructions for students


Recent Submissions

Now showing 1 - 20 of 1137
  • Item
    Autonomous UAV Swarms: Distributed Microservices, Heterogeneous Swarms, and Zoom Maneuvers
    (The Ohio State University, 2024-05) Angueira Irizarry, Kevyn; Stewart, Christopher
    Over the last few years, precision agriculture has greatly benefited from advancements in Unmanned Aerial Vehicles (UAVs). UAVs used in crop map ping allow farmers and researchers to tailor farming practices to the specific needs of individual management zones. Yet despite their benefits, traditional exhaustive approaches to UAV remote sensing are constrained by the low battery capacities of UAVs. To optimize battery usage, we present our work across 3 papers on alternative reinforcement learning (RL) and multi agent reinforcement learning (MARL) approaches to UAV remote sensing, namely: SoftwarePilot 2.0, heterogeneous swarms, and the Zoom maneuver. Starting with SoftwarePilot 2.0., we introduce a software package that sup ports scalable autonomous UAV swarms through microservice model design, container deployment technologies, and specialized MARL policies. These improvements to SoftwarePilot 2.0. reduced energy costs by 50% and improved swarm decision-making by 2.1 times from base SoftwarePilot. Moreover, we further explored multi-agent strategies through the extrapolation of multiple health metrics with heterogeneous UAV swarms. Heterogeneous swarms can capture data from multiple types of sensors, e.g. RGB, thermal, multi-spectral, and hyper-spectral cameras, and then extrapolate for various distinct health metrics across the whole field. Our preliminary results showed 90% accuracy from extrapolation from sampling only 40% of the field. Lastly, we proposed a study on the battery and accuracy tradeoffs of Zoom maneuvers. Moreover, Zoom maneuvers or changes in altitude trade battery for increased local accuracy. Our study considers the computational battery cost and flight battery cost tradeoffs of autonomous vs. RL implementations of Zoom maneuvers. Ultimately, this paper provides new insights into the execution and performance of various autonomous and multi-agent UAV remote search strategies.
  • Item
    The Effect of Electrostatics on the Trajectory and Adhesion of Dust Particles in Jet Engines
    (The Ohio State University, 2024-05) Sabau, Tanner; Bons, Jeffrey
    Despite the continued advancements in jet turbine engines to be more efficient and run hotter, foreign object damage (FOD) still poses a large problem and can negate some of the improvements. Extensive research has been done on how FOD such as dust particles can impact cooling holes and build up within engines over time, but there has been limited research on the influence of electrostatic forces on these particles. This research aimed to investigate how electrostatics from the atmosphere and tribocharging affect the trajectory and adhesion of dust particles within jet turbine engines. Three tests were conducted: a vertical drop test within a uniform electric field, an angled drop test on a charged plate, and an impingement cooling deposition test using a charged target. In the vertical drop test, 0-20µm Arizona Road Dust (ARD) was dropped through an electric field 90cm tall and 10cm apart. Mass fractions were collected on different sections of the electric field to determine alterations in the particles’ trajectory. The angled drop demonstrated the influence of charge on the tested dust’s adhesion strength by dropping the 0-20µm ARD on a 10° tilted, 90cm tall, charged plate. The impingement cooling deposition test simulated engine conditions, with 0-10µm ARD blown through a narrow tube at high velocity and temperature impacting a charged target. The results indicated that at low speeds the electric field significantly influenced the ARD trajectory, with smaller particles more affected than larger particles. Increased voltage in the angled drop test correlated with larger deposits at the top of the plate, which showed higher adhesion strength of the dust. Surprisingly, higher plate voltage in the impingement cooling deposition test led to decreased dust accumulation in the region of interest. By exploring dust behavior within engines under electrostatic influence, these findings will open new possibilities for enhanced engine performance and durability.
  • Item
    Metabolic Cost of Overground Walking Using a Motorized Walker-like Exoskeleton
    (The Ohio State University, 2024-05) Hu, Jasper; Srinivasan, Manoj
    Those with movement disorders commonly use exoskeletons and assistive walking devices. The purpose of most exoskeletons and assistive walking devices is to reduce the effort required to perform a task by providing mechanical assistance to the user. A way to measure effort is by studying metabolic cost. Thus, by providing mechanical assistance when performing a task, the metabolic cost of performing the task may decrease. Previous studies showed a passive walker-like exoskeleton cart in some configurations decreased the metabolic cost of walking [1]. Since such findings, a brushless DC motor was added to the cart to provide an adjustable assistive forward force on the user while walking. This research aims to determine the relationship between the intensity of the assistive forward force, metabolic cost, and walking speed when overground walking with the cart. To do so, metabolic and walking speed data was collected from five subjects using a COSMED K5 system and a video camera respectively. Each subject completed ten six-minute trials where nine of the trials were used to collect walking data. The trial order was randomized. One of the trials was used to collect normal walking data in which the subjects walked normally, without the cart. The metabolic cost and walking speed from this normal walking trial were used as the baseline metabolic cost and walking speed during data analysis. When averaging across the five subjects, the cart was shown to decrease walking speed for all but the greatest assistance level (potentiometer setting 4.2, corresponding to large assistance). The cart increased the average metabolic cost per unit time from normal walking for all assistance levels except one (potentiometer setting 3.6, corresponding to medium assistance) and increased the average metabolic cost per unit distance from normal walking for all assistance levels. However, analyzing the data subject-wise, we found that the cart decreased metabolic rate below normal walking for four out of the five subjects and that the cart decreased the metabolic cost per unit distance for three out of the five subjects (with one subject showing a marginal increase). These decreases suggest that the cart may be promising. Further research will be required to determine how the cart can be modified or how much training the subjects need to be provided such that the cart becomes a more uniformly assistive walking device. Should the cart be modified and proven to reduce the metabolic cost of walking, the cart may replace other devices and provide a framework for future device designs.
  • Item
    Illuminating Neural Pathways: Modulating Motor Cortex and Related Pathways with Photobiomodulation
    (The Ohio State University, 2024-05) Yu, Emily; Saygin, Zeynep
    Parkinson’s Disease (PD) and Major Depressive Disorder (MDD) are prevalent disorders of the brain which may benefit from treatment via nonpharmacological neuromodulation. Current approaches include transcranial magnetic stimulation (TMS) which has been shown to noninvasively stimulate the brain with transient effects. However, the effects of TMS are limited to superficial regions. A newer noninvasive method that can reach subcortical structures is transcranial photobiomodulation (tPBM), which uses visible to near-infrared light to stimulate neurons. In this work, we have built a multi-wavelength tPBM system to assess the effects of light of 1064 nm wavelength on brain activity and connectivity. Given that motor skills are affected by MDD and PD, we focused on stimulating the primary motor cortex (M1). Healthy adult volunteers (N=10) underwent four sessions of stimulation consisting of active tPBM and TMS along with respective sham sessions. Structural and task fMRI were taken before and after stimulation. Participants performed a motor task to explore the effect of stimulation within each individual’s motor cortices. Here we report that tPBM successfully stimulated M1 to a similar amount of activation as TMS in a safe and short-term manner. The effects of tPBM and TMS stimulation on motor activation was not significantly different within participants, suggesting that the two types of stimulation induce similar effects on the somatomotor cortex. We also observed significant changes in functional and structural connectivity between the motor cortex and thalamus after PBM stimulation. These results are in line with previous literature. Interestingly, TMS stimulation did not seem to affect structural and functional connectivity. Our preliminary results reveal that tPBM can cause changes in activation and connectivity of the somatomotor cortex with short-term effects that are comparable to the gold standard of neuromodulation, TMS. tPBM is still a relatively novel technique with more to be explored. PBM harnesses the neuroprotective properties of brain cells and may be useful when pharmacological options are unsuccessful or cannot be tolerated. Compared to other stimulation methods, PBM can penetrate further into the brain, providing more options for treatments of mental illnesses.
  • Item
    Development of Cathode Catalysis for Co-Electrolysis of CO2 and H2O
    (The Ohio State University, 2024-05) Lvovich, William; Ozkan, Umit
    The increase in CO2 emissions and the strained future energy capabilities of fossil fuels have driven research to focus on sustainable energy. To address these alarming issues, high-temperature co-electrolysis of CO2 and H2O conducted within a solid oxide electrolysis cell (SOEC) can be utilized. It not only reduces the amount of CO2 in the atmosphere, but also generates usable synthesis gas (H2 and CO). This synthesis gas can be further processed into light hydrocarbons via Fischer-Tropsch synthesis. Therefore, it is evident that improving the SOEC is of upmost importance. In this study, the double perovskite cathode catalyst Sr_(2-x)Fe_(1.55)Mo_(0.45)O_(6) (SFM) will be investigated for its high electronic and ionic conductivity, which is ideal for the co-electrolysis of CO2 and H2O at high temperatures. The study will highlight various permutations of the base SFM via A-site deficiency and nickel and cobalt doping, which improve the electrical conductivity and the chemical activity. The proposed study will centralize on developing a cost-effective, chemically active catalyst for the aforementioned electrolysis reactions to ensure SOEC’s will become a viable economic competitor in the marketplace.
  • Item
    Numerical Investigation and Optimization of a Miniature In-Pipe Hydropower Turbine
    (The Ohio State University, 2024-05) Lenk, Charlie; Belloni, Clarissa
    Pico hydropower energy generation, an emerging field in renewable technology, has faced limited funding and attention in the past due to the inherent lower efficiencies found for smaller scale turbines. This research addresses the need to provide a small amount of power in situations where consistent power availability and water quality may be a concern. Specifically, this research numerically investigates the performance of miniature turbine designs suitable for integration within water piping of facilities or homes to provide enough wattage for UV LED water disinfection at the point of use (POU) scale. Aiming to maximize the energy extracted from water flowing at POU water pipe pressure and flow rates, this study conducts Computational Fluid Dynamics (CFD) analysis. This research evaluates the hydrodynamic performance of a pre-manufactured turbine. After this, changes to the nozzle geometry were made in an attempt to improve efficiency. After varying the height of the nozzle, a max hydraulic efficiency of 22.05% was found at a flow rate of 2.2 GPM and a runner rotational velocity of 2200 RPM. The results suggest that further numerical analysis of various changes to the turbine geometry would be useful. A more complete nozzle geometry parameterization study is recommended as the next optimization step. Changing the runner hub size, and the number of blades on the runner is another recommended course of action. Once a promising novel geometry has been identified, the Hydro and Aero Energy Group (HAEG) will continue with producing a physical prototype of a POU miniature hydropower turbine.
  • Item
    Developing a Sensor Calibration Device for Sensory Brain Computer Interfaces
    (The Ohio State University, 2024-05) Terveer, Michael; Lucas, Timothy
    Brain Computer Interfaces (BCIs) provide individuals with severe neurological conditions the ability to regain functions they have lost, such as movement and speech. Significant research with these BCI systems has been performed specifically in spinal cord injury patients in an effort to create methods for restoring motor functions to patients. A lack of knowledge, however, exists in creating BCI systems that can naturally restore sensory function in such patients. New implantable fingertip force sensors have been developed that could provide patients the ability to regain this crucial function of the nervous system. These novel sensors, however, require calibration with commercial sensors to be effective and useful for different patients who require their own unique stimulation parameters. This project aimed to create a first-generation (“alpha”) prototype of a calibration apparatus for these sensors for specific patient scenarios as well as test and validate the prototype in a real patient setting. Following the design, prototyping, and validation process, a modular, custom forearm cradle device was created and proven to be capable of registering force outputs in a clinically relevant setting. A production process was created along with the first iteration of the cradle, permitting the expansion and tailoring of this process for usage in several different patients. Furthermore, the function of the device in registering the force outputs of patients’ hands was verified in four healthy volunteer subjects. This was accomplished using the commercial sensors selected for verifying the custom sensors against. Results showed that force outputs could be distinguished based on the time duration of force applied, the magnitude of strength applied, the location of force application, and ultimately the individual applying the force. These initial results confirm the need for the patient-specific calibration setup that has been developed in this study.
  • Item
    High Temperature Oxidation of Refractory High Entropy Alloys (RHEAs) and the Use of High Entropy Rare Earth Oxide (HERO) Coatings
    (The Ohio State University, 2024-05) Marino, Isabella; Locke, Jenifer
    Refractory high entropy alloys (RHEAs) possess favorable material properties, such as high melting points, resistance to corrosion, microstructural stability, and mechanical strength, which makes them a strong candidate for aerospace, automotive, and nuclear applications. To ensure safe high temperature performance, understanding and minimizing the rate of oxygen diffusion, and specifically high temperature oxidation, in RHEAs is imperative. The inherent high temperature oxidation performance of two RHEA compositions provided by the United States Air Force Research Laboratory are being investigated, after varying high temperature exposures in a controlled oxygen environment for various time intervals. Additionally, the effects of surrounding each alloy in a high entropy rare earth oxide (HERO) sintered powder sheath provided by the University of Virginia is investigated under identical conditions. Dilatometry is used to compare thermal expansion between the substrates and HERO coating. Thermogravimetric analysis is used to measure oxidation resistance and SEM and X-ray techniques are used to understand the microstructure.
  • Item
    A Techno-Economic Analysis of Lithium-Ion and Sodium-Ion Battery Energy Storage Systems in Buildings with Renewable Energy Sources
    (The Ohio State University, 2024-05) Casey, Austin; D'Arpino, Matilde
    Battery energy storage systems (BESS) provide various benefits, including renewable energy integration and grid support ancillary services, such as frequency regulation. The most commonly dispatched BESS are those of Lithium-ion (Li-ion) chemistries. However, concerns relating to depleting Lithium reserves as well as ethical issues relating to the extraction of Lithium and associated materials have provided motivation to search for a sustainable battery chemistry that can provide comparable, if not better, performance than Li-ion. An emerging technology of interest that may address this issue is the Sodium-ion (Na-ion) battery, due to the vast abundance of Sodium in nature and theoretical performance being similar to its Li-ion counterpart. In this Thesis, a techno-economic analysis of Li-ion and Na-ion BESS have been performed to determine their performance and financial metrics. The Energy Advancement and Innovation Center (EAIC) has been used to model a building with renewable energy resources, and a BESS of each chemistry has been designed to meet the optimal dispatch strategy to meet the load demand of the EAIC as determined by the Renewable Energy Integration and Optimization Tool (REopt) provided by the National Renewable Energy Laboratory (NREL). The financial metrics of each BESS have been obtained using the cashflows from REopt to determine the net present value (NPV) of each project. Then, using the power profile from REopt as an input, an electrothermal simulation for grid-connected storage has been developed to determine the performance metrics of each BESS chemistry. To simulate the performance metrics for frequency regulation, a RegD signal from PJM was used as the power profile for each BESS. The financial metrics of each BESS performing frequency regulation were obtained by calculating the credits using data from PJM. Finally, a comprehensive comparison of each technology has been provided for each BESS chemistry, accounting for its performance and financial metrics, while also considering the social and environmental impact of each technology.
  • Item
    Changes in biophysical properties of collagen hydrogels by pH for tumor modeling
    (The Ohio State University, 2024-05) Schrieber, Grant; Song, Jonathan
    The role the extracellular matrix (ECM) plays in cancer invasion and metastasis is not well defined and has become a central focus of cancer researchers. Collagen-based hydrogels have been primarily used to reconstruct the ECM for cancer studies as they recreate a 3-D model representative of native tissue observed in vivo. However, due to the heterogeneity of solid tumors (such as breast cancer), this has made our ability to recapitulate tumor microenvironments (TME) difficult. This set of studies aimed to use gelation-pH of collagen-based hydrogel systems as a mechanism to manipulate the biophysical properties of the in vitro models. Through changes in gelation-pH, we achieved a two fold increase in stiffness from acidic to basic conditions. In addition, collagen polymerization dynamics were distinctly different based on gelation-pH. The ECM characterization methods elucidate the importance of gelation-pH on the biophysical properties of the in vitro models and provides a manner to represent a wide range of tissues from the body. Future studies will look to correlate the current stiffness and polymerization kinetic results with the microarchitecture of the hydrogels to better understand the biological phenomenon within these models.
  • Item
    Experimental Investigation of a Miniature In-Pipe HydroTurbine
    (The Ohio State University, 2024-05) Abdulrahman, Farah; Belloni, Clarissa
    A recent report published by the UN revealed that 26% of the world’s population lacks access to safe drinking water, and 46% lacks access to basic sanitation. These groups cannot afford to revamp their infrastructure and filter their water. This project provides a cheaper alternative by disinfecting water at the Point-of-Use scale by employing pico-hydropower turbines to power UV-LEDs, disinfecting the water as it runs out of the faucet. Available options had previously been found to have turbine efficiencies of <1%. Previous research showed that an increase to approximately 10%-20% efficiency should successfully power the UV-LED device. The current research entailed the testing, analysis, and optimization of two available off-the-shelf options. The testing was done by running tap water through the turbines and measuring the current and voltage using an oscilloscope, as well as manually recording the rotational velocity and flow rate. This was done at several resistances and pressures. These measurements were then used to calculate power and efficiency. The turbines tended to fail sporadically at higher pressures or after several consecutive trials. To counter this, they were taken apart and lubricated before being used for testing again. The findings showed a moderate increase in efficiency to approximately 8% when operating with resistances around 220 Ohms. However, after months of being taken apart and used for testing, the turbines reached the end of their lifecycle, yielding efficiencies of 2% or less despite being lubricated. Further examination of the turbines shows that water damage due to poor casing design is the likely culprit of the low power output. After testing the generator, it was determined that the generator was up to 50% efficient, with most tests yielding results between 20%-45%. This suggests that the turbine is most likely the limiting factor preventing the system from functioning at full capacity, as opposed to the generator. The next steps include redesigning the turbine and the casing in order to reduce the likelihood of water damage and prolong the lifecycle of the turbine. With continued testing, the 10% efficiency threshold could be met and potentially power the UV-LED device to disinfect the water.
  • Item
    Defining mechanisms of binding utilized by the infective endocarditis pathogen Streptococcus mitis
    (The Ohio State University, 2024-05) Narayana, Anupama; King, Samantha; Wood, David
    Streptococcus mitis is an oral commensal and a common cause of subacute infective endocarditis (IE), a bacterial infection exhibiting an upward trend in mortality and hospitalizations. Subacute IE often occurs when viridans group streptococci, commonly found in the oral cavity, enter the bloodstream and bind to damaged heart endothelium. These bacteria specifically bind to platelets, via sialic acid, that accumulate at the site of damage. Platelet binding is a key step in the development of IE and a potential therapeutic target. Until recently all described mechanisms of sialic acid binding were mediated by serine rich repeat proteins (SRRPs) such as Fap1 produced by Streptococcus oralis. However, our group identified a novel sialic acid-binding protein, AsaA, in S. oralis strains that lack Fap1, suggesting additional mechanisms of sialic acid binding may exist. Although S. mitis is a common cause of IE, no mechanisms of adhesion have been defined. To start filling this gap in knowledge, we examined the distribution of orthologs of fap1 and asaA across 20 genome sequenced S. mitis strains. What was initially thought to be an ortholog of asaA, was actually two distinct genes producing putative adhesins, AsaA and AsaB, differing in the frequency of repeat domains and genomic location A fourth putative sialic acid-binding adhesin, Rib, containing a different repeat domain, was identified by predicted structural conservation of sialic acid-binding domains. Further analysis suggests asaA and rib evolved from asaB. While the genes encoding MonX and AsaA are mutually exclusive in both S. oralis and S. mitis, some strains are predicted to produce one of these adhesins and AsaB and/or Rib. These data suggest some S. mitis strains produce novel sialic acid-binding adhesins and multiple adhesins may contribute to platelet binding. To start investigating how S. mitis binds sialic acid, we selected B6, which contains asaB and monX and efficiently binds sialic acid on platelets. We have generated a monX mutant and the construct to inactivate asaB. Once all mutants have been obtained, platelet binding assays with single and double mutants will be performed. The long-term goal of these studies is to gain a comprehensive understanding of all mechanisms of platelet binding. As binding is a key step in IE pathogenesis, our hope is that these studies will allow the development of broadly effective therapeutic strategies.
  • Item
    Developing a 3D Metastasis Model for Metastatic Progression in Medullary Thyroid Cancer
    (The Ohio State University, 2024-05) Lewis, Mitchell; Dedhia, Priya; Skardal, Aleksander
    Thyroid cancer is currently the fastest-growing cancer in the United States. Medullary thyroid cancer (MTC) originates from the parafollicular cells, otherwise known as C cells, of the thyroid gland and has a significantly worse prognosis than all other thyroid cancers. The majority of MTC patients present with advanced disease, already having either regional or distant metastases. Unfortunately, currently, existing treatments cannot improve most MTC patients’ prognosis in these advanced disease states. Therefore, this study aims to create and validate a 3D model capable of identifying the underlying signaling pathways that govern metastatic progression with the hope of discovering potential novel therapies for MTC in future applications. To test this model, we will 1) perform LIVE/DEAD assays to validate cell viability and 2) model MTC metastasis using cell injection metastasis-on-a-chip (MOC) designs. Early results show promise of a new invasion model of MTC for therapeutic targets.
  • Item
    Numerical Analysis of the Impact of Debris Protection on Hydrokinetic Rotor
    (The Ohio State University, 2024-05) Frear, Keegan; Belloni, Clarissa; Niebuhr, Chantel
    Global energy consumption is rising each year, and with an increase in energy consumption comes an increase in the reliance in renewable energy. Currently, hydropower is one of the most prominent forms of renewable energy used across the globe. However, the implementation of hydropower facilities is expensive, time consuming, and comes with a litany of environmental issues. An alternative to hydropower is hydrokinetic energy; hydrokinetic energy harnesses kinetic energy from moving streams of water rather than relying on the potential energy difference between water bodies. The purpose of this research is to validate a hydrokinetic turbine using Computational Fluid Dynamics (CFD). Previously the turbine was analyzed using Blade Element Momentum theory (BEM). However, this is a simple approach that does not account for turbulence created by the blades. StarCCM+ is utilized to construct the CFD model, which is then validated against BEM-derived flow parameters for maximum power efficiency. Validation of the BEM results using CFD is crucial because it facilitates further testing of additional geometrical features necessary for real world implementation. Additionally, this study investigates the impact of cylindrical rods placed upstream to act as a grid to block debris. Determining the grids optimal positioning based on their effect on the turbine’s power efficiency.
  • Item
    Developing 3D Lung Rendering Algorithms for Computational Modeling of Gene Therapy Drug-Delivery Routes for ARDS and Ventilator-Induced Lung Injury.
    (The Ohio State University, 2024-05) Nkwocha, Chikamnele; Ghadiali, Samir
    Acute Respiratory Distress Syndrome (ARDS) is a severe condition defined by oxygen depletion and pulmonary edema. Mechanical ventilation, a common care method, can further exacerbate the damage created by ARDS by increasing inflammation and mechanical stress within the lungs. Thankfully, gene therapy technology has shown promise in mitigating inflammatory responses caused by ventilator-induced lung injury. Gene therapy for ARDS relies on successfully delivering aerosolized microRNA nanoparticles (which include miR-146a) to the injury site. The use of 3D computational models of damaged and healthy CT lung scans can help realize the effective drug-delivery route of miR-146a while minimizing pulmonary fluid occlusion. This project explores the use of three distinct algorithms for developing computationally compatible 3D lung models: active contours segmentation, trained neural network segmentation, and 3D spline lofting.
  • Item
    Human Activity Recognition via Garment-embedded EMG Sensors
    (The Ohio State University, 2024-05) Goetz, Danny; Srinivasan, Manoj
    Human activity recognition (HAR) is a fast-growing field due to the increased desire to include this technology in many commercial electronic devices and its applicability to healthcare. The most common approach to HAR is to train a machine learning model with sensor data from human activity. Currently, accelerometers, inertial sensors, and gyroscopic sensors are the most common devices used to collect data for HAR. Another type of sensor that can provide data on human activity is an electromyography (EMG) sensor. EMG measures the electric signals due to person’s muscles activation. This thesis explores the possibility of using EMG sensors for HAR. To investigate this possibility, garments with embedded EMG sensors from ExoForce. The garments, when worn by participants, are designed to measure the activation of triceps, biceps, and calf muscles. The study consisted of participants who were asked to do a series of defined exercises (pushups, jumping jacks, curls, etc.) while wearing the EMG garments. A subset of the data from the participants was used to train a multinomial logistic regression machine learning model for classification. The rest of the data from the participants was used to test the accuracy of the model. The accuracy of the model was 45.31% on the test data (much higher than random guessing, which would be 12.5%), 98.44% on the training data, and 71.88% on the whole data set. The results from this study show that EMG sensors can be used as the basis for a device that accurately identifies various human activities, while future work could focus on improving accuracy and exploring specific applications. This finding could lead the way to improved commercial or medical devices aimed at HAR.
  • Item
    Continuous Stairs Climbing Control Design for Bipedal Robots
    (The Ohio State University, 2024-05) Wang, Boyuan; Hereid, Ayonga
    While robotics technology advances and more real-world applications are used in the industry, legged robots, unlike robotic arms or wheeled robots, are not widely utilized despite their various benefits, such as improved movement performance in complex terrains. This could be due to several limitations in both hardware and software. This paper focuses on developing a reliable and efficient control algorithm for bipedal robots walking on long, continuous stairs. This enhancement aims to increase their movability in real-world scenarios, fully utilizing their advantages of better operation and relatively less space requirement in complex terrain compared to traditional wheeled robots. To enable bipedal robots to walk on continuous stairs, the paper proposes improvements to the traditional Angular Momentum Linear Inverted Pendulum (ALIP) planner. These improvements enable the 2D planar five-link walker robot, RABBIT, to handle vertical and horizontal displacement of the center of mass (COM) and to calculate foot displacement as a trajectory in the simulator. The findings suggest that the modified traditional dynamical model for bipedal robots requires fewer resources to generate control algorithms. As the working environment, stairs, a reparative environment, this offers an advantage with utilizing Bezier approximation to regulate the swing-foot and COM displacement trajectory onto the bipedal model itself. This work highlights the advantage from model-based approach that it takes less time and inputs to generate the control algorithms and the transparency during the simulation comparing to the black box-like process of the learning-based approaches, but with an inherent limitation of lacking versatility. And for LIP model specifically, without introducing other policies or methodologies, it is insufficient to handle the vertical displacement of the COM for bipedal robots.
  • Item
    Advancing Incremental Profile Forming: Experimental Analysis of Axial Grooving and Deep Learning Modeling
    (The Ohio State University, 2024-05) Campbell, Ian; Seetharaman, Satyanarayana; Srinivasan, Krishnaswamy
    This thesis presents the experimental analysis of the axial grooving process during Incremental Profile Forming (IPF), a manufacturing process developed at TU Dortmund, Germany [1]. The IPF machine has advanced machine control but limited capability to accommodate certain process mechanics like springback, resulting in inaccurate part geometry [2]. A computationally efficient model of axial grooving is sought that can be used for process control. The goals of this thesis are to obtain a better understanding of the axial grooving process and to investigate the efficacy of modeling deformation using Deep Learning (DL), which has been shown to be an efficient modeling tool [3]. Axial grooving processes were performed on the IPF machine, and geometry data collected. The experimental process was then simulated using the finite element method (FEM), and resulting geometry data was compared to measured data to gain insight on the nature of deformation, the accuracy of the FEM model, and the accuracy of the IPF sensors. Experimental results show that laser line optical sensors are accurate to within the desired error threshold of 0.1 mm, however the current coordinate transformation and alignment procedure results in an error of 0.18 mm. Geometric data shows uneven groove depth between the entry, steady-state, and exit regions of the grooves. Tool type was shown to have an effect on overall groove geometry. For the DL investigation, a 2D beam bending problem [4] was modeled using both data driven neural networks and Physics Informed Neural Networks (PINNs). A study was done to understand the effect of input normalization on model accuracy. The PINN predictions were compared to data-driven network model outputs and the known analytical solutions. A second study was performed to evaluate the effect of input domain size on PINN model accuracy. It was determined that input normalization allows data-driven models to handle vast input domains but introduces error in PINN models. From the domain study, it was shown that PINNs are more accurate if their input domains are restricted. This research will advance understanding of IPF process mechanics including errors involved in on-line sensing. Furthermore, the DL study has identified key limitations of using PINNs for modeling problems with large input and output domains.
  • Item
    Design, Prototyping, and Evaluation of Two Dexterous Multi-Segment Pneumatic Soft Robot Grippers
    (The Ohio State University, 2024-05) Yan, Chenyang; Su, Haijun
    Soft robot grippers have become a popular research topic in recent years. Unlike traditional mechanical grippers, soft robot grippers utilize the deformable properties of their soft materials to envelop target objects through deformation, thus achieving grasping suitable for various shapes. Moreover, robots made of soft materials can enhance the safety of interactions with humans. This paper will introduce two prototypes of pneumatic soft robotic gripper designs. The first prototype features a gripper with two air chambers, while the second prototype has three air chambers. The paper aims to detail the design and fabrication process of two multi-segmented fingers and focuses on evaluating the performance of the two-segment gripper. The evaluation includes conducting flexibility tests in three distinct ways: testing the curvature of the grippers under different input pressures, assessing stiffness variations, and controlling the pressure difference in each air chamber to conduct a grasping adaptability test. The results showed that this two-segment soft finger prototype exhibited strong adaptability in grasping items of varying shapes and sizes and demonstrated a high degree of flexibility when continuously rotating objects. Each segment was able to withstand a maximum pressure of 25 psi and execute bending commands in less than 1 second. In summary, this study explores the complementary balance between the fabrication of multi-air chamber soft robotic fingers and their flexibility and grasping strength. The study further illustrates that through optimized design, soft robotic fingers can perform complex operating tasks without the need for sensors and complex control systems. This further illustrates the ease of operation of soft robotic fingers in daily life and is a step closer to widespread application in the future.
  • Item
    Simulation Design of 4D printing of thermally-activated Shape-shifting Structures
    (The Ohio State University, 2024-05) Kong, Xiangyuan; Sutradhar, Alok
    4D printing is one of the latest manufacturing techniques. This is a technology for the creation of objects that evolve in shape and structure over time in response to external stimuli such as heat, magnetic fields, and light. Building upon the foundation of 3D printing, this technology relies on smart materials such as shape memory polymers (SMPs) to facilitate these transformations. SMP is a highly promising and extensively utilized material in the field of 4D printing. This is due to its wherein the remarkable shape memory effect (SME). The experiment design done by lab is printing the shape-shifting parts of SMPs. The activation process begins when microwave radiation is applied to the printed structure, warming the integrated heater elements, which subsequently raise the temperature of the adjacent shape memory polymers (SMPs). To accomplish the results by simulation, Finite Element Analysis (FEA) is introduced. By using the FEA methods, the model of this shape-shifting design should be built to get the results close to the experiment, and it can also be used to optimize such design in the future. By analyzing simulation results, we investigate deformations caused by microwave radiation and compare them to experimental data. In this study, by replicating existing theories and models to learn the process of building accurate simulations, a shape-shifting simulation is created to mimic the experiment results.