Interdisciplinary – Page 24 – UROP Spring Symposium 2021

Interdisciplinary

Powertrain Strategies for the 21st Century: The Tesla Effect

For the Tesla research project, I, along with my mentor Bruce Belzowski and research assistant Daniel Nemmert, helped formulate and evaluate the results of a survey that asked automotive industry experts including executives, managers, from auto manufacturers, suppliers, as well as academics, consultants, government officials, and NGOs for their opinions on Tesla’s unique technology, distribution methods, and company strategies. The survey also inquires about Tesla’s future global expansion and market share. Tesla’s recent rise and success within the exclusive electric automotive industry coupled with their unorthodox approach and advanced technologies make them a potential game-changer within the industry. Additionally, the world’s seemingly inevitable shift to electric vehicles, a sector in which Tesla is a global leader, makes them all the more relevant. This critical transition period to electric vehicles poses a problem for automotive companies tasked with undergoing a major operational and technological shift into the new market. Thus, this survey, which will be distributed in February and analyzed in March, will be used to further evaluate which of Tesla’s unique company characteristics are their strengths and weaknesses and to establish predictions regarding the future of the electric vehicle market including major players and imminent advances. The survey results will give insight that will inform competing automotive companies about other methods that, based on the methods’ successes, can assist and guide them in how they conduct their business in the future. Overall, this research will help automotive companies prepare for and adapt to consumer needs and preferences more effectively, creating more efficient businesses and better customer experiences.

Powertrain Strategies for the 21st Century: The Tesla Effect Read More »

Literature Review of Human Factors Armored Vehicle Research Conducted by the United States Army

The topic of this research is driver user interfaces for autonomous Army vehicles and the associated workload, usability, and other related factors. The Army has done experiments on autonomous and tele-operational vehicles and their interfaces for the last 20 years, without any overarching studies to display the results across all of the experiments. The question this project is answering is: what are the important trends and general direction of this Army work on autonomous and tele-operational vehicles over the last 20 years? The purpose of the project is to go through 2 decades of Army literature and experiment reports and write literature reviews to summarize them. After finishing the literature reviews the important information will be condensed into tables to create a method to compare and contrast the findings of each report in a way that has not been done yet. This research is being done by reading Army Reports and writing literature reviews to summarize the important information. These reviews will be analyzed by creating summary tables encompassing all of the findings from the reports that have been read so that similarities, differences, and trends can be identified. The information is still being condensed into summary tables and are not yet ready to identify conclusions and trends. However, when conclusions are found, they will likely identify trends in soldier preferences, where the army should go next, and what systems are more and less successful, and how that can influence future decisions.

Literature Review of Human Factors Armored Vehicle Research Conducted by the United States Army Read More »

Literature Review of Human Factors Armored Vehicle Research Conducted by the United States Army

The topic of this research is driver user interfaces for autonomous Army vehicles and the associated workload, usability, and other related factors. The Army has done experiments on autonomous and tele-operational vehicles and their interfaces for the last 20 years, without any overarching studies to display the results across all of the experiments. The question this project is answering is: what are the important trends and general direction of this Army work on autonomous and tele-operational vehicles over the last 20 years? The purpose of the project is to go through 2 decades of Army literature and experiment reports and write literature reviews to summarize them. After finishing the literature reviews the important information will be condensed into tables to create a method to compare and contrast the findings of each report in a way that has not been done yet. This research is being done by reading Army Reports and writing literature reviews to summarize the important information. These reviews will be analyzed by creating summary tables encompassing all of the findings from the reports that have been read so that similarities, differences, and trends can be identified. The information is still being condensed into summary tables and are not yet ready to identify conclusions and trends. However, when conclusions are found, they will likely identify trends in soldier preferences, where the army should go next, and what systems are more and less successful, and how that can influence future decisions.

Literature Review of Human Factors Armored Vehicle Research Conducted by the United States Army Read More »

Next Generation Combat Vehicles: Safety and Usability

The U.S. Army has been conducted research on advanced technology for several decades, summarized in about 30 reports and presentations. A major theme has been the use of automation and new technology to support the operation of ground vehicles, both manned, remotely operated, and autonomous. Topics have included partial automation of the driving task, various designs for displays to present information to drivers and commanders (e.g., head-mounted displays), the design of controls used for driving, requirements for training, the effect of degraded visual environments on driving, automatic target recognition, and a range of other topics. The presentation will provide an overview of the research methods and tools used, as well as a summary of the findings.

Next Generation Combat Vehicles: Safety and Usability Read More »

Global Electric Vehicle Strategies

Battery electric vehicle (BEV) sales have continued to break records in 2020. However, the progress does not seem to translate evenly across all vehicle segments. In this research, two sets of data are collected and analyzed. First, historical and current BEV models are sorted into segments (model data). Second, model specifications such as range, MSRP (manufacturer suggested retail price) and battery size are collected for each model across different years (specification data). The analysis focuses on the US, Chinese and European markets, and 2018 is taken as the base year. Based on the model data, the number of existing and new models in each segment is steadily increasing. However, the large and full-size vehicle segments are lagging behind. In the specification data, performance and price display a direct relationship and controlling for price reveals similar per unit performance increases across segments. We conclude with a brief comparison to the internal combustion engine (ICE) vehicle segments. As an ICE vehicle becomes larger, fuel economy and price actually changes inversely. Manufacturers benefit from minor functionality and comfort upgrades in larger and more expensive ICE vehicles. In conclusion, a healthier growth in the large and full-size vehicle segments is needed to support the momentum for electric vehicles. This means the industry will probably adjust its production to provide greater price differentiation, especially in the large and full-size segments, as the EV industry matures.

Global Electric Vehicle Strategies Read More »

Progress Towards Developing a Quantitative Definition for Atmospheric Rivers

Atmospheric Rivers (ARs) are channels of water vapor in the atmosphere that result in a large amount of precipitation. They cause over 1.1 billion dollars of damage annually in the United States. As a result, accurately interpreting the presence of an AR has become crucial to ensuring the safety of citizens, especially those who live in regions frequently impacted by ARs. Progress is difficult, however, because the accepted AR definition is largely qualitative, and researchers who have tracked ARs use different identification methods. Therefore, an analysis of the strengths and weaknesses of various detection algorithms is of great interest. The Atmospheric River Tracking Method Intercomparison Project (ARTMIP) is a multicentered, collaborative effort to quantify the differences among AR detection techniques. This study focuses on eight AR detection algorithms run on ECMWF’s ERA5 reanalysis product. The time intervals among the ERA5 datasets range in hours and the time frame generally covers the years 1980-2019. To facilitate comparison among detection approaches, we compared transects of AR landfall from East Asian countries using the Python Programming language. Assuming that each algorithm identifies the presence of an AR perfectly, the transect plots should be identical among the eight datasets. However, some plots differed significantly from the others. For future studies, one should develop a standardized approach to detecting and interpreting ARs. An accurate interpretation of these weather systems will not only save lives but help countries economically.

Progress Towards Developing a Quantitative Definition for Atmospheric Rivers Read More »

Investigation of South America Atmospheric Rivers

Previous studies have identified that atmospheric features with low-level moisture transport, known as atmospheric rivers (ARs), are connected to extreme precipitation events around the globe. While these ARs provide vital water resources to communities, they are also known to cause fatalities resulting from flooding and landslides. The severity of their impacts is expected to increase with climate change due to increased atmospheric moisture. Our understanding of what drives changes in AR behavior is still incomplete. While there has been a wide breadth of research conducted on ARs in the North Pacific region, much work has yet to be done in order to fill gaps in knowledge about ARs across the globe. The investigation of South America ARs aims to quantify how our understanding of ARs depends on algorithm choice. This specific study focuses on detection algorithms run on JRA-55 reanalysis (55 km resolution). The dataset consists of six different algorithms, namely the ARConnect, GuanWaliser, IDL, Mundhenk, Payne and Reid algorithms. Python is used for intercomparison and visual representation of the differences and strength of various detection algorithms, focusing on the understudied region of southern South America (15°N-60°S, 110°W-16°W). This project aims to identify algorithms that perform poorly and common AR characteristics where there is agreement. In doing so, it is the hope that affected communities across this region will better be able to respond to incoming extreme weather associated with AR events.

Investigation of South America Atmospheric Rivers Read More »

Analysis of the Human Ovarian Cell Landscape using Histological and Immunofluorescent Staining

Improvements in cancer treatments have led to the increase of cancer survival rates in recent decades. While these treatments are life-saving, they have cytotoxic effects on the ovaries, thus depleting the ovarian follicle supply. The number of follicles in the ovaries is nonrenewable, leading to premature ovarian insufficiency (POI). In vitro follicle culture could serve as a broad fertility preservation option for these patients, but much is still unknown about the mechanisms driving early follicle development. Single cell sequencing of ovarian tissue from deceased donors can be used to characterize the role of stromal cells in follicle development as well as transcriptional differences between follicles at the different stages of development. To verify the cell populations identified using single cell sequencing we will use hematoxylin and eosin (H&E) and immunofluorescent staining to characterize the spatial landscape of follicles and stroma in donor ovarian tissue. I will quantify follicles at different stages of development across donors and use immunofluorescent staining to validate and support the sequencing data. This work will deepen our understanding of follicle development and supporting cell types, leading to development of human follicle culture systems and a broad fertility preservation option for cancer survivors.

Analysis of the Human Ovarian Cell Landscape using Histological and Immunofluorescent Staining Read More »

Machine Learning Methods for Robust ECG Beat Detection

This research was conducted to determine what types of machine learning algorithms were best to determine whether an individual is experiencing heart arrhythmias—such as atrial fibrillation, a type of arrhythmia that can lead to a number of fatal conditions such as blood clots, stroke, and heart failure. Using Python and the sklearn library, a number of machine learning models were tested for accuracy of ECG peak detection, which included Logistic Regression, Linear Discriminant Analysis, K-Nearest Neighbors, Classification and Regression Trees, Gaussian Naive Bayes, and Support Vector Machines. The accuracies varied based on the proportion of noise within the ECG file, so an introspective algorithm was developed to choose the optimal peak detection algorithm based on the estimated noise level measured in an ECG. This can be used in the future as a convenient way to accurately determine the presence of heart arrhythmias using wearable devices such as a smart watch.

Machine Learning Methods for Robust ECG Beat Detection Read More »

Machine Learning in Cardiovascular Medicine

Atrial fibrillation is the most common type of heart arrythmia in humans, and our goal is to develop a method of characterizing medical data signals to assist cardiologists to better manage atrial fibrillation. Predicting atrial fibrillation from ECG signals because ECG patterns may change from patient to patient, and include noise. In our study, we provide a comparison across different existing methods of classifying ECG time series using machine learning models. We tested two different machine learning models: a Fully Connected Neural Network and a Support Vector Machine, and compared their performance and accuracy on the Physionet Atrial Fibrillation Challenge Dataset (www.physionet.org). We found that although a Fully Connected Neural Network as a deep learning model is pretty robust when given large amounts of data, the Support Vector Machine method performed better than deep learning when limited data is available to use. Potential applications of this include live analysis of ECG signals that could assist cardiologists to perform diagnosis in real time.

Machine Learning in Cardiovascular Medicine Read More »

lsa logoum logo