Toyota Financial Services (TFS), being part of a highly regulated industry needs to ensure that all risk management, governance process, and controls are in place to ensure compliance. This entails documenting all the business processes, definitions of their data elements, connecting the defined data elements to the physical attributes in their various applications and databases. Furthermore, they need to document the lineage of the data to ensure that it is flowing correctly through their ecosystem. In addition to these, they must ensure the data quality at the source and through the transformations, it goes through while flowing in their ecosystem.
The problem of the disjointed system to record, store, check and correct all the data in the ecosystem/ no holistic view of data is affecting the employers/ business partners of TFS, the impact of which is unorganized data, manual process of linking physical and business data elements which is time-consuming.
A solution that our team is working towards is to build a data portal where data will be organized by business areas such as Loan Originations, Insurance, Servicing, etc., and various classifications under those areas. We are also implementing a google-like search for any data element which would bring up business definitions, physical attributes, data quality rules, profiles, and any related data associated with it.

Toyota Financial Services (TFS) is the largest automotive lender in the nation with over $125B in total assets. TFS offers financing, leasing, protection plans, and other financial services to customers and dealers all across the United States. In order to serve their customers better, it was required to have a comprehensive view of the customer. The TFS Customer 360 team has worked with TFS to create a Customer 360 view by harnessing the power of graph databases, semantic queries, and graph visualization tools. This view represents all direct and indirect relationships that exist for a customer and will be made available to different stakeholders in the company to make more informed decisions and to better identify potential opportunities.

Oxidative stress refers to the imbalance between free radical activity and antioxidant activity in the body, and is known to play a crucial role in diseases such as age-related macular degeneration in eyes and various neurodegenerative diseases (Alzheimer’s and Parkinson’s). To help the body target and rebalance this process, the Green group at TCU has developed pyridinophane macrocycle frameworks (PyN3, Py2N2) for the development of a small multimodal molecule with direct targeting of oxidative stress through various approaches (metal binding, N-oxide formation, radical scavenging, and Nrf2 pathway activation). The group proposed a library of ligands as modifications to the pyridinophane frameworks to enhance antioxidant activity, which resulted in 18,000 possible molecule structures. Computational pre-screening will be essential to select the most promising candidates for synthesis and experimental tests. We wrote a program in Python using the open-source RDKit toolkit to generate a library of 13,000 prospective reduced-dimension pyridinophane macrocycle derivatives from SMILES strings based on the variation of ligands and attaching position to the frameworks, screen these compounds for their basic chemical and pharmacological properties, and identify those that fit the required biocompatibility, metabolic stability, and permeability for medicinal drug development. The properties to be computed through the virtual screening are molecular weight (MW), solubility, ring count, Lipinski’s parameters for orally active drugs, which includes octanol-water partition coefficient cLogP, number of hydrogen bond donors (HBD) and acceptors (HBA), and polar surface area (PSA). This program, therefore, helps save time and resources for synthesis while offering better optimization of chemical frameworks, and thus it can be applied to the development of various types of medicinal drugs.

The Monnig comprises a scientific research collection (the Monnig Meteorite Collection) as well as a Museum (the so-called “Gallery”). The exhibit has an educational game and interactive video screens. While there have been some updates to the technology in the exhibits in the last twenty years, most of the Gallery remains unchanged since its opening in 2003.

The current design is not inclusive for visually impaired visitors. contents are not accessible for non-English speaking visitors, and only less than 5 % of the 3000 meteorite collections are displayed. We address these problems by developing a tour assistance application with sufficient accommodations for visually impaired visitors using Android tablets that will be provided by the MMG to its incoming visitors with a capability of being fine-tuned to the individual’s preference.

The application begins with three separate menus, each with its own screen, which allows the user to customize the app to their needs. Menu 1 allows them to select their language. The beta version includes English, Spanish, and French, but later versions could consist of more language options. Menu 2 provides a font size selection, and menu 3 allows users to identify their color blindness type. These variables can be reset or changed at any time. The Monnig Meteorite Collection database has images of all of the meteorites within the Gallery and the text from each exhibit will be reproduced within the application. The images will be shown on a high contrast background (as compared to the exhibits) to allow for better viewing of the samples. Voice transcription will also eventually be available. Wayfinding, the ability for the user to identify where they are within the Gallery, on the application will be achieved in one of two ways. QR codes will be placed on each display case, allowing users to scan and locate the relevant exhibit on the map when needed. In addition, Bluetooth receivers will be used so that the application can identify where in the Gallery the user is located.

Chinese Learning Platform is a part of STARTALK Program – a federal grant program funded by the National Security Agency. The mission of the program is to assist students in learning languages identified critical by STARTALK, including Chinese, Arabic, Hindi, Persian, Korean, Russian, and Turkish. Our project aims to support students in learning Chinese, and will be extended to other languages in a near future. The supports include, but are not limited to, assistance in vocabulary, listening, reading, writing, and speaking Chinese. In addition, our project contains well-designed functionalities dedicated to language learning, thus further improving the learning experience for students.

DC to DC Conversion is important in modern electronics, and to the automotive industry. It is the process of converting a direct current (i.e constant) signal into another form of direct current (DC). A small-scale example of this is a car adapter, which converts the 12 volts provided by a car outlet into the 5 volts a cell phone needs to charge, known as a ‘step down’ converter. The main objective of the project is to design and test a bidirectional DC to DC conversion system. Most DCDC converters available on the market are unidirectional, i.e., either ‘step down’ or ‘step up’ the DC signal. Those that can switch are called bidirectional converters, but many available cannot handle the higher requirements of an electric motor. A system that can switch directions based on specific system parameters allows for situational flexibility, and the use of new devices for more efficient energy use. The supercapacitor is one such device. They provide power more efficiently than batteries but can only store a small amount of energy. They must be recharged often, which requires a step-down conversion from a power source (much like the car adapter example). Supplying the motor from supercapacitors requires a step-up conversion. So, to use, and reap the most benefit from these supercapacitors, switching from step up to step down based on their charge is a requirement. Ultimately, this would allow for the use of supercapacitors in an EV as part of a future project.

In this presentation, the process of creating a map of an area using ground truth data will be explained. The overall objective of this research project is to be able to capture a remote image of a land mass and be able to discern what sections within that image are a certain material. This is done through the matching of spectral signatures, which are unique for every physical material found on earth. A demonstration of spectral signature matching will be shown to understand the basic idea of how the mapping is done. A model expanding on this idea with the use of ground truth data will be shown with results showing how the map will be made.

Hydrodynamic Analysis of Surfboard Fin Performance set out to analyze how the outline and size of a surfboard fin can impact performance. Performance analysis involved running images from the manufacturer’s website through a MATLAB code that would process the image to determine an appropriate, numerical method, based on fluid dynamics, to explain categorical differences between fins. After testing for differences between categories for the following performance metrics: the vertical line of action, the horizontal line of action, the ratio between the tip area and the rest of the fin, and the resulting angle created by comparing the vertical and horizontal lines of action, the angle was found to be the most statistically significant factor for determining fin categories. Moving forward, users can input an image, along with the fin dimensions, to determine the performance characteristics of a fin, without having to purchase a fin. This project explains the underlying equations that are utilized, the fundamental assumptions that are made, how the results are generated, and how users can interpret the results.

Autonomous drones have been commercially available for decades. The integration of sensors has allowed robots to interact with their environment and resulting in autonomy. This quadcopter team takes on the challenge of creating an autonomous quadcopter using a frame, motors, electronic speed controllers, propellers, a Raspberry Pi, and an RPLidar. The team achieved remote control flight of the drone through pre-installed software—QGroundControl. The onboard computer will collect data using the RPLidar sensor and then send the data to the flight controller. Setting the robot (talker) and the virtual machine (listener) as nodes, they can communicate with each other through the ROS master.

Generative design implementation in this project had the goal of replacing sheet metal structures previously used to hold relays and electromechanic switches with 3D printed structures. The generative design software has the benefit of minimizing the mass of the structure, while keeping its structural integrity. The software does this by iterating through designs solving for stresses at each step, deciding where it is better to place a structure and then cutting mass at points where the structural integrity would not be compromised. Although the software creates a design on its own the user must define certain parameters: the preserve geometry (fundamental geometry for operation), obstacle geometry (sections that the software should leave without obstruction), the expected load case, manufacturing method, and material to be used. The end result is that the computer creates most efficient parts, allowing for a plastic 3D printed part to be able to safely replace one made of metal.