Organizational Unit:

Undergraduate Research Opportunities Program

Permanent Link

https://hdl.handle.net/1853/70802

Parent Organization

Organizational Unit

Office of Undergraduate Education

ArchiveSpace Name Record

https://finding-aids.library.gatech.edu/agents/corporate_entities/1180

Full item page

Publication Search Results

Now showing 1 - 10 of 16

Seeing Red: American Tourism to the Eastern Bloc, 1960-1975

(Georgia Institute of Technology, 2019-05) Haskin, Kayleigh Georgina

Theoretical literature asserts that tourism should lead to better interactions between nations with different ideas and cultures. However, empirical studies find that this is often not the case, and certain pre-trip factors are more influential in changing tourists’ opinions than the experience itself. This study examines one of these potential factors: the role that the news media plays in shaping public opinion about foreign countries prior to travel. Using a case study of American tourists to the Eastern Bloc from 1960-1975, this paper suggests that media portrayal contributed to the negative views Americans held of the Soviet Union and the lack of opinion change after travel. Using the counterexample of Hungary, this paper also suggests that this portrayal was unique to the Soviet Union, and not reflective of the Eastern Bloc as a whole. Finally, it offers a potential new avenue for future research on opinion change in tourists—the consideration of pre-trip domestic factors, such as the news media and the overarching geopolitical context.
3D Reconstruction of Live Chickens in Poultry Houses

(Georgia Institute of Technology, 2018-05) Muni, Aneri Dhirendra

Poultry houses require daily monitoring to ensure animal health and proper house operation. One task involves observing the average growth rate of the house to adjust the daily feed. In addition to being labor intensive and time consuming, it is difficult for the farm owners to find consistent labor to fill these jobs. This project looks at the possibility of estimating the weight of a chicken based on the volume of the chicken as captured by a 3D model. We present a system capable of reconstructing dynamic scenes, i.e. chickens in a poultry house, by fusing together depth scans captured using a Microsoft Kinect. Like DynamicFusion, our approach involves discretizing the live depth frame into nodes and estimating individual 6D transformations before fusing them together to reconstruct scene geometry. This approach doesn’t use any prior model of template, making it applicable to a wide range of dynamic objects and scenes.
Reviewing the Effects of Poverty and Food Scarcity on the Prevalence of Type II Diabetes Mellitus in the Metro Atlanta Area

(Georgia Institute of Technology, 2017-08) Sledge, Kyle Emerson

Type II diabetes is a medical condition that is prominent in both society and medical research. In the metro Atlanta region, over nine percent of the population has contracted the disease. It is not a random disorder but the result of individual factors and local environments. One of the most influential factors is diet, which is directly impacted by what foods are available locally. Healthy food acquisition can be almost impossible in food deserts, or areas of the country that contain either no or extremely limited locations to purchase healthy food. These deserts litter the metropolitan area of Atlanta, and are often intermixed inside more rural or low-income areas. This paper identifies these Atlanta food deserts and analyzes them for a correlation to the prevalence of type II diabetes. The data confirmed my hypothesis that there were correlations between these clearly identifiable sites in greater Atlanta and elevated levels of incidence of type II diabetes. Clayton and DeKalb counties had the largest proportions of their populations inside food deserts at 45.1% and 21.8% respectively, and had increased percentages of their population with type II diabetes: Clayton County surveyed with 10.6% of its population having the disorder, and 8.9% of DeKalb’s inhabitants had acquired the disease at some point. The data suggest that food deserts have adversely affected the health of thousands of people. On the other hand, there are sectors that can be targeted and changed to directly increase the quality of living for over 17% of the metro Atlanta populace that are currently living inside of them.
Biorobotic Locomotion: Biped Humanoid Walking using Optimal Control

(Georgia Institute of Technology, 2017-05) Bindhi, Malavika

This paper explores the use of optimal control for quasi-static bipedal walking trajectory synthesis. Optimal control differs from the traditional method of inverse kinematics used for generating trajectories. On setting up an optimal control problem and solving it, a set of joint angle trajectories for executing a portion of the gait can be synthesized using the forward kinematic description, as opposed to the method of inverse kinematics. Although optimal control solution requires the same computational processing, the advantage is the simplification of problem set up procedure. This research aims at developing a quasi-static walking gait for a humanoid by defining a set of optimal control problems with constraints and costs defined as functions of joint angles. OPTRAGEN, a Matlab toolbox, is used to convert the optimal control problem to a non-linear problem which is then solved by a numerical optimization program. The goal is to generate a planar walking gait by setting up an optimal control problem that minimizes the change in joint angles while keeping the body stable. The generated planar gait was tested on the physical robot. On completion of this, the same methods were also used to generate trajectories for a 3-Dimensional quasi-static walking gait.
Localization of Electromagnetic Sources in Complex Processors

(Georgia Institute of Technology, 2017-05) Pollmann, Eric

Recent advances in computer hardware security research have shown that electronic devices are vulnerable to electromagnetic (EM) side-channel leakage. An attacker can steal sensitive information from an electronic device that is not connected to the network just by measuring the electromagnetic emanations produced by the device. The feasibility of a successful EM side-channel attack has grown significantly in recent years with the reduction in cost of electronic test equipment and the proliferation of electronic devices. Smartcards, cellphones, and laptops can be susceptible to side-channel attacks, and with the trend towards internet-of-things (IoT) and millions of connected devices, the side-channel will become of increasing security importance. While most of the current literature in electromagnetic side-channels focuses on demonstrating novel attack methods on various devices, relatively little research focuses on characterizing the EM emanations. Research presented in this thesis aims to develop a metric to measure the side-channel energy per instruction available to an attacker. The importance of this is to understand the effect of instruction level events (i.e. ADD, LOAD, NOP) on the electromagnetic emanations from a cellphone. Understanding the correlation between the instruction run and the corresponding electromagnetic footprint can help to understand which operations cause the greatest amount of electromagnetic leakage as well as provide a basis of measurement. In addition, a method of localization is proposed to identify the source of the EM emanations. A magnetic dipole equivalent source model is proposed and measurement results are compared with simulation results of the model. These results aim to provide a deeper insight into the nature of electromagnetic side-channels as well as provide hardware and software designers with the information needed to combat unnecessary EM emanations.
System Dynamics-Based Mapping for Closed Loop Control

(Georgia Institute of Technology, 2017-05) Dixit, Anushri C.

This paper focuses on obstacle avoidance using the walking gait and vision. Inspiration for the gaits was drawn from behaviors found in nature as well as prior contributions in the field of robotic locomotion. A quadrupedal robotic platform was designed and fabricated to support these studies and experiments. The walking gait was implemented on the platform using inverse kinematics and a map was developed connecting the system dynamics to the extrinsic control parameters, namely, the stride length of the robot and the turn of each leg. The paper has implications in path planning for bio-inspired robots in rough terrains. The goal of the research is the synthesis and evaluation of increasingly dynamic quadrupedal locomotion gaits like walking, trotting, and hopping for navigation over unknown terrain.
Designing an In-Home Scalable Robotic Arm Exoskeleton for Hand Rehabilitation Therapy

(Georgia Institute of Technology, 2016-07-18) Tuck, Jonathan

Cerebral palsy and stroke are both debilitating neurological disorders that may permanently damage a person’s body movements and muscle coordination. Recent methods to help facilitate therapy in cerebral palsy patients and stroke survivors include the patient interacting with a robotic arm exoskeleton that is connected to a tablet gaming suite, which has shown to encourage the patient to complete his or her rehabilitation regiment. A major problem currently facing robotic arm exoskeleton rehabilitation therapy is that commercially available robotic arm exoskeletons are relatively bulky, heavy, expensive, and typically fit only one size, limiting the amount of patients who can benefit from this therapy. This paper describes a method to create an easily scalable robotic arm exoskeleton for use in hand rehabilitation therapy, through 3D printing. This paper then illustrates a study where the developed robotic arm exoskeleton was compared to a commercially available robotic arm exoskeleton in order to determine efficacy.
Localizing Embeddings for Recommendation Systems using Binary Pairwise Comparisons

(Georgia Institute of Technology, 2016-07-18) Oshaughnessy, Matthew R.

Recommendation systems predict the preferences of users (who may be, for example, customers of an online shopping website or moviegoers), to various items (for example, consumer products or movies). One way this information is modeled is the ideal point model of preference, in which items and users live in an n-dimensional Euclidean space where each dimension represents an attribute and a small distance between an item and user indicates the user has a preference for that item. We seek to determine an embedding of many items and users given only binary pairwise comparisons of the form "user x prefers item q_i to item q_j." First, we present an optimization-based framework for localizing new items and users given an existing embedding. We demonstrate that user localization can be formulated as a simple constrained quadratic program. Further, we show that although item localization produces a quadratically-constrained quadratic program which is difficult to solve, we can make the problem more computationally tractable by strategically combining comparisons to make the quadratic constraints into linear constraints. Finally, we show that by iteratively applying this localization method to every item and user, we can recover an embedding that agrees with almost every comparison, allowing us to iteratively improve the accuracy of a noisy embedding or even create an embedding using no a priori knowledge apart from the list of pairwise comparisons. Throughout, we present implementation details and optimization algorithms which make the recommendation system computationally efficient even with large datasets and dimensionalities.
Enhancing Audio Perception in Noisy Environment Audio Signal Processing

(Georgia Institute of Technology, 2016-07-18) Rawat, Vasundhara

Enhancing audio perception in noisy environments is currently one of the most explored topics in the field of signal processing. After thorough research, engineers who have developed noise suppression filter using filter banks or any related approaches are faced with the difficulty of removing the noise caused when the input sound signal is passed through a filter. This paper introduces a technique using ERB (Equivalent Rectangular Bandwidth) that enhances the attributes of a clean signal such that it sounds clearer in the presence of background noise. The intended audiences are scientists and researchers in the field of exploring audio signal processing.
The Closed Loop Optimization of Deep Brain Stimulation Programming

(Georgia Institute of Technology, 2016-07-18) Singh, Ravinderjit

Deep brain stimulation (DBS) is a procedure used to treat movement disorders such as Parkinson's disease. The current procedure for programming the parameters for DBS is time consuming and prone to error. The DBS programming procedure can be significantly improved using a closed-loop optimization approach. Due to recent advances in quantitative assessment metrics, the capability to translate a closed-loop optimization procedure for DBS programming from simulation to clinic has become more possible. Previous literature has presented closed-loop approaches that utilize evolutionary algorithms. It is very difficult to implement an evolutionary algorithm in the clinic because they typically require a large number of parameter evaluations. A parameter evaluation is testing how well a certain set of DBS parameters work. It is difficult to do a large number of parameter evaluations due to time constraints and patient fatigue. A response surface based closed-loop optimization approach for DBS programming is presented that has higher potential to be translated to the clinic because it requires much less parameter evaluations.