Computer Science Program 2023 Projects

Aerospace Dilution of Precision Automation

Faculty Advisor: Zilong Ye
Client: Aerospace Corporation
Liaisons: Andre Chen, Denny Ly, Karina Martinez, Pablo Settecase

Students: Richard Bailon, Xico Blanco, Yuridia Ginez, Nathan Gonzales, Andrew Jarmin, William Leung, Pedro Eduardo Ramirez, Cesar Salazar, Aaron Simental, Scott Yadong Sun

Aerospace needed a solution for visualizing the Dilution of Precision (DOP) data without needing subject matter experts. DOP is a measure of how well a group of GPS satellites' signal strength is, and Aerospace Corporation currently has the capability to perform this visualization. However, their process is manual and time-consuming, requiring their subject matter experts to create correctly formatted files to be analyzed by a program that produces a user interface with a visualization of the DOP values.

The manual pipeline currently in place at Aerospace Corporation requires significant human intervention, which results in slow turnaround times and a high potential for error. In addition, their visualization tools are antiquated and require significant processing power, further slowing down the process.

The team of engineers tasked with solving this problem is creating an automated solution that can take the raw satellite data, analyze it, and produce a user-friendly visualization without needing expert intervention. The goal is to automate the process, reduce errors, and create a more efficient and user-friendly system for a broader audience.

The engineers are exploring various technologies and solutions to help them achieve this goal. The hope is that by automating the process and implementing more modern tools, Aerospace Corporation can significantly improve its visualization capabilities and better serve its clients.

View Aerospace Poster

BOE Sidewalk Assessment System

Faculty Advisor: Jungsoo (Soo) Lim
Client: Dept. of Public Works, Bureau of Engineering, LA City
Liaisons: Ted Allen, Alisa Blake, Miguel Grajeda, Raul Virgen

Students: Gerardo Bracamontes, Joshua Chang, Christopher Chau, Robert Cueto, Salvador Gomez, Zachary Justus, Ricardo Miranda, Bryan Nguyen, Jorge Sanchez, Amy Torres

This is the fifth term of a multi-year project. In the last term, a rover has been successfully fabricated. Now, the rover is capable of 1) moving autonomously, 2) measuring crossing slopes and running slopes, 3) collecting GPS data, and 4) taking photo images. In this term, we will continue developing various software by focusing on the following tasks: the students shall work with Civil Engineering Department to verify the correct procedure for collecting slope measurements in the current data collection process, and they shall port the current program developed on Leo Rover to a new rover platform by making the necessary modification; they will develop a strategy to collect horizontal/vertical displacement on the sidewalk by equipping the rover with needed components. (i.e., distance measure lidar, sonar, etc.), and develop software to display geographically referenced data as needed for Sensing & Robotics for Infrusture (SRI) lab.

View BOE Poster

Faculty Advisor: Jungsoo (Soo) Lim
Client: Santa Barbara Public Defender's Office
Liaisons: Deepak Budwani

Students: Mario Avila, Jimmy Castillo, Joseph Comeaux, Sherina Marquez, Ethan Ngo, Alexis Ponce, Karen Quan, Rosa Robles, Alexander Voisan, James Yokley

Currently, the Public Defender's office has a content management system called eDefender, parent company Journal Technologies. The Department has transitioned to a fully paperless case management system. As the transition from paper to paperless continues, additional server space will be needed to host expanded growth and to maintain our digital client files.Box.com is a cloud content management tool that would allow the Public Defender to be compliant with CJIS/HIPAA requirements, have available storage to store all files digitally, have a collaborative platform, workflow automation, build API's to eDefender and allow governance. The Department currently has approximately 80+ terabytes of case data it needs to transition to the cloud; yearly consumption is twenty-four (24) terabytes and increasing exponentially yearly. The Public Defender plans to leverage Box.com to build the necessary infrastructure and support our network needs. The project will migrate the entire infrastructure for the Public Defender's office to the cloud with Box.com. We would like to build automation, including integrating with AWS and Microsoft Cognitive skills to build facial recognition and transcription (similar to LA County Public Defender project). The goal is to build a speech recognition machine-learning algorithm to build transcripts of any audio or video users feed into its web-based interface. Once files are uploaded, they would be translated, transcribed, facial recognition scanned and keyword searched and placed the text in transcript templates. Once uploaded, tags (911 calls, body-worn camera, interview) would be added to trigger the Box Skills. Once the process is complete, Box.com will integrate with eDefender to trigger notifications, tasks, and alerts upon transfer. Minimum metadata will be sent to eDefender to trigger automated notifications (disc number, date received, case number, file format, file size, type of media (photos, 911 calls, body-worn camera, videos, interviews, audios, forensics, Internet of Things (i.e., ring camera)).

Phase two of this project will look to OCR discovery pdfs (police reports) to extract party information from the report and send that extracted metadata to eDefender through an API. This will greatly reduce the time it would take to add parties to the case. A detailed process flow chart is attached to the email as well. Another phase to explore is whether discovery can be redacted based on pre-set conditions.

View Box Poster

Digitization and Modernization of PDHelpdesk Ticketing System

Faculty Advisor: Chengyu Sun
Client: Santa Barbara Public Defender's Office
Liaisons: Deepak Budwani, Brent Modell, Luis Ramirez

Students: Christian Armendariz, Hoai Cao, Brandon Estrada, Gilbert Hopkins, Geovanny Huerta, Marie Karibyan, Nshan Kazaryan, Biruk Mengeste, Mark Perez, Kevin Trochez Grajeda

Santa Barbara Public Defender's Office uses a homebuilt, Outlook-based IT ticketing system which receives and stores IT requests submitted by Public Defender staff. This system is functional but limited in its capabilities. It outgrew its original purpose and operating parameters. The system does not allow for advanced data analytics or workflow automation and prohibits IT staff from obtaining a comprehensive understanding of the systems and practices used within the workspace. In this project, we have built a dedicated ticketing system that provides the necessary issue monitoring and resolution tools to assist IT employees in performing their duties. Moreover, the system offers an intuitive interface that facilitates seamless communication between technicians and customers, enabling them to collaborate on resolving issues and accessing the comprehensive history of previous tickets. This system helps to enhance customer service by simplifying the process of handling requests, efficiently resolving issues, and ensuring that no customer needs are overlooked.

View Helpdesk Poster

Firefox Desktop Browser: Improving Import Usability for Users

Faculty Advisor: Chengyu Sun
Client: Mozilla Foundation
Liaisons: Katherine Patenio, Niklas Baumgardner, Mike Conley, Neil Deakin, Gijs Kruitbosch

Students: Zachariah Harris, Nolan Ishii, Alvin Lew, Evan Liang, Bryan Macoy, Ani Movsesian, Brian Pham, Angel Villalobos, Steven Wang, Portia Wu

Firefox allows users to import their data, such as bookmarks and passwords, from other browsers. The existing importer is functional but lacks some important features, does not support newer browsers, and has an outdated and clunky user interface. Our project improves the importer by supporting more data (e.g., favicons) and browser types (e.g., Opera, OperaGX, Vivaldi, and Chrome as a Snap package). We also implemented a new import process and modern UI that greatly simplifies the steps for users. These changes will slowly be released in Firefox to the public, enhancing the browser's overall user experience.

View Firefox Poster

Genetic Factors Determining COVID-19 Susceptibility and Severity

Faculty Advisor: Navid Amini
Client: Program for Education and Research in Biotechnology (CSUPERB)
Liaisons: Haley Kirk

Students: Riese Atianzar, Rohan Chatterjee, Francisco Contreras, Emily Gonzalez, Luis Gonzalez, Ting Fung Ha, Juan Hernandez, Carlos Hernandez, Jimuel Julaton, Chen-Ching Lin

Our goal in this project is to determine clinical factors of a poor prognosis in patients with COVID-19 infection. We will examine medical records of patients with COVID-19 infections confirmed by polymerase chain reaction (PCR). Logistic multivariate regression models adjusted for age and sex will be constructed to analyze independent predictive factors associated with death, ICU admission, and hospitalization of these patients. As part of the project, we are looking at the link between COVID-19 severity (e.g., COVID-19-related death) and the blood type of patients. This topic has been a controversial topic in the research community.

View COVID Poster

JPL 1 - Trek Virtual Reality (VR) Room

Faculty Advisor: David Krum
Client: Jet Propulsion Laboratory (JPL)
Liaisons: Eddie Arevalo

Students: Lucca Andrade Guedes Galvao Coutinho, Fabio Carrasco, Enrique Guardado, Ruben Heredia, Ari Jasko, Bryan Lopez, Ly Jacky Nhiayi, Ayush Singh, Rizwan Vazifdar, Justin Vuong

Solar System Treks is a JPL web portal that provides NASA data from various planets, asteroids, and moons. This data is used by scientists, research partners, and the public, including students and teachers. The data includes satellite photography and 3D terrain models. Collaborative visualization, especially in virtual reality (VR) and augmented reality (AR), can help generate new insights. A team of researchers or students can examine the data together and collectively generate and discuss new ideas and hypotheses. A prototype of VR capability called "Trek VR Room" was developed by a CSULA graduate student for the Solar System Treks Project. It is a VR client which connects to Trek back-end servers to display Mars data contents based on Mars Trek (https://trek.nasa.gov/mars). This is built on SteamVR using Unity and only works with HTC VIVE headsets. It allows users to search, view and manage Mars data layers. It provides simple tools for distance measurement and elevation profile generation. It also includes a tool that supports highlighting and spotlighting an area for a detailed examination of the data.

We propose a project for next year's CSULA student capstone team to upgrade the prototype with the latest toolkit to support multiple VR devices by updating or rebuilding the Trek VR Room prototype with Unity and OpenXR to support both Oculus products, including Oculus Quest 2 and HTC VIVE headsets. JPL will provide VR devices for the student team to use for their implementation. The student capstone team will be able to analyze and learn from the prototype and work with JPL Solar System Treks team members to design and implement the modification needed for the upgrade. After this initial upgrade, later development cycles could focus on making it functional for different planetary bodies starting with the Moon. This project would create useful tools supporting the collaborative visualization of NASA solar system data. This research and development would also enable a variety of future projects that could examine a variety of interesting topics involving virtual reality, augmented reality, visualization, and analytics.

View JPL1 Poster

JPL 2 - MoonTrek Augmented Reality (AR) for Lunar Telescope

Faculty Advisor: Weronika Cwir
Client: Jet Propulsion Laboratory (JPL)
Liaisons: Natalie Gallegos & Shan Malhotra

Students: Nadir Abdusemed, Jackson Bentley, Jesus Cruz, Youssef Elzein, Derek Guevara, Joe Hineno, Rich Ho, Owen Ramirez, Salman Sheikh, Alex Sherzai

Previously, the Cal State Senior Design Capstone Project focused on lunar image processing - particularly, coregistration between telescope images and LRO WAC images. Some work has been done to render 3D Sun, Earth, and Moon geometry and Earth-Moon visibility. The project's goal is to build upon the previous year's project, but the focus is on AR.

For phase one, students will work with a telescope electronic eye-piece that will provide students with access to real-time images. The effort will focus on using the co-registration algorithm and allow real-time AR metadata to be added to the image. The metadata will be queried from a Trek web service.

Phase two will focus on "super zoom." A telescope image will be co-registered with a WAC mosaic. Users will be able to zoom into their image - as they zoom past the resolution of their image, pixels from the LRO cameras will be used. This will also be supplemented with metadata. The experience will allow backyard space enthusiasts to use their home equipment to connect what they see with NASA's high-resolution datasets.

Phase three will build on the 3-D modeling that will allow a user to see the relative position of the Earth, Moon, and Sun as it is in real time from their current Earth location.

View JPL2 Poster

MathWorks Sensor Fusion for Autonomous Systems

Faculty Advisor: Manveen Kaur
Client: MathWorks
Liaisons: Sumit Tandon (Mathworks representative for Cal State system)

Students: Hagop Arabian, Daniel Gallegos, Roberto Garcia, Gerardo Ibarra, David Neilsen, Patrick Sangalang, Jonathan Santos, Deepanker Seth, Angel Tinajero, Xiao Wang

The goal of this project is to develop a sensor fusion algorithm for vehicle pose estimation using AI-based techniques. Vehicle pose refers to a vehicle's position and direction. Vehicles typically employ a variety of onboard sensors, such as cameras and LiDAR sensors, to gather environmental information and estimate vehicle pose. Traditionally, pose estimation used classic filtering techniques such as Kalman Filtering. However, we are now exploring Machine Learning (ML) and Artificial Intelligence (AI) based methods to help us with estimation. In this project, we will design a software application that can fuse vehicle sensor data from multiple sensors and use ML to analyze the data and estimate a vehicle's pose. General components of the work to be done include:

• Identifying a limited set of sensor types and data that we consider most relevant to vehicle pose estimation.
• Identifying an appropriate ML method through a survey of recent literature and industry practices.
• Training and testing of the ML algorithm.
• Revision of software application with other appropriate sensor data. The idea behind this step is to determine if additional sensor data adds value to our vehicle pose estimation method.

View MathWorks Poster

RoboSub

Faculty Advisor: Richard Cross
Client: RoboSub
Liaisons: Mark Tufenkjian

Students: Thomas Benson, David Camacho, Bailey Canham, Brandon Cao, Roberto Hernandez, Andrew Heusser, Hector Mora-silva, Bart Rando, Victor Solis, Milca Ucelo Paiz

The RoboSub Senior Design project is a joint project between a team of Electrical/Mechanical engineering students and a team of Computer Science students. The purpose of the project is to design, build, and program a fully autonomous underwater vehicle (AUV) that will compete in the international RoboSub competition hosted by Robonation. The computer science team will design and implement the software that pilots the Autonomous Underwater Vehicle (AUV), providing navigation, stabilization, object detection, and task-handling capabilities. This software will provide the AUV the ability to detect objects and obstacles using image recognition and then respond to this information to maneuver itself and interact with its environment to meet the objectives outlined in the competition's rules. A team of Mechanical and Electrical engineering students will be designing and building the AUV intended for use in the competition.

The International RoboSub competition is hosted by Robonation and is held annually at the end of July at the US Navy's Transducer Evaluation Center (TRANSDEC) in San Diego. The competition normally consists of a set of obstacles arranged in the TRANSDEC pool that each team's submarine must navigate autonomously. This year the competition will be held July 25-31, 2022.

View RoboSub Poster

SAYA.life User Portal

Faculty Advisor: Manveen Kaur
Client: SAYA.life
Liaisons: Sanjay Poojari

Students: Esteban Alvarado, James Bodden III, Christopher Case, Bryant Garcia, Marian Gomez, Alvin Hong, Fidel Izquierdo Jr, Arthur Keshishyan, Han Linwu, Cristian Moreno

The purpose of this project is a restyling of the current SAYA.life web application, emphasizing modularity and reformatting of embedded analytics. The application features both a tenant view, where individuals can view the water consumption of their apartment/home, and some forecasting to see how their water consumption affects their billing. More importantly, for Saya, they requested a landlord view, with the ability to view all property's water usage, send messages to tenants, and monitor analytics on water consumption broken down by the individual unit. Our liaison with Saya specified the project will be done using React, and many of the other details have been left up to us. We plan to provide them with a sleek, intuitive web app design and some very sound scaffolding for them to build out on if they so desire.

Fall 2022:

Create wireframes for the pages and brainstorm ideas for how the page should look for various users. Create user profiles, covering a wide range of familiarity with web applications to ensure the broadest spectrum of people possible can use our project. Gain familiarity with their API and build some of the supporting infrastructure we will need for our project (database for testing back-end functionality and forecasting).

Spring 2023:

Implement the website with a focus on the landlord view. We are making sure the application is easy to navigate and allows the user to customize the Dashboard to their liking. Provide some embedded, real-time forecasting, with the ability for follow-on teams to add more as Saya sees fit.

View SAYA Poster

Security and Privacy Protection on IoT Devices

Faculty Advisor: Huiping Guo
Client: CS Department and LAX Engineer
Liaisons: Anson Fong

Students: Ricardo Castellanos, Luis Cortez, Diego Flores, Adrian Holloway, Jivan Karapetian, Minsu Lee, Asahel Monroy, Ashley Pablo - Ramirez, Garrett Stewart, Guang Wu

HOSS is a web application that scans a network for all connected devices. HOSS will utilize the NMAP API to scan and retrieve the list of devices within the user's network. After retrieving the list, HOSS will cross-reference its own database with historical data regarding each device's security/privacy settings. After HOSS cross-references the database, it compiles and aggregates two security scores. One score is for the privacy of each listed device, and the other is for the security of each device. These scores are provided to the user, along with a summary of why each device received the score along with recommendations on improving the security/privacy of each device which was flagged with concerns.

View IoT Poster

Self-Attention Mechanism of ChatGPT

Faculty Advisor: Yuqing Zhu
Client: Computer Science
Liaisons: Russ Abbott

Students: Fardeen Abir, Nathan Campos, Anthony Edeza, Jose Flores De Santiago, Edward Kim, Kenneth Lieu, Kevin Mateo, Michael Nguyen, Roberto Reyes, Maggie Yang

AI systems have been studied and applied in small-scale applications for decades. ChatGPT, however, has taken over the world with its almost human-like replies and fast processing of inputs from humans. This kind of breakthrough in AI technology, along with his ease of access to the general public, has opened the doorway to more similar systems to be developed. Our team has focused our attention on Andrej Karpathy's small-scale gpt program and has turned our efforts into analyzing the core learning mechanism of ChatGPT. Because ChatGPT is a fairly new application of the GPT language model, this project will dive into the core learning mechanism called self-attention and break down the process step by step.

The goal of this project is to explore the use of self-attention using small-scale data with characters to replicate the format of various texts. The primary deliverables will be (A) an expository report on how self-attention works.

View ChatGPT Poster

Human Perpetual Models - Small Unit Drone Optimization (SUDO)

Faculty Advisor: Elaine Kang, David Krum
Client: Army Research Lab
Liaisons: Paul Fedele, Eric Holder

Students: Jonathan Aguirre, Lloyd Castro, Jean Espinosa, Peter Han, George Hernandez, Hugo Izquierdo, Raymond Martinez, Bruck Negash, Jesus Perez Arias

This project aims to digitize human audio and visual perception models amalgamated through decades of study by the Army Research Lab and transform those models into a dynamic, easy-to-use, and modern programming language. Our program takes into account factors such as environmental noise, a person's hearing threshold, atmospheric conditions, effects of ground composition, temperature, and 1/3 octave bands (among others) throughout multiple frequencies and tracks the effects they have on auditory propagation loss. This will lead to an output that can determine a person's auditory detection range of any inputted target hardware.

This newly-developed program will then be used to create a testing simulation in a 3D environment using the Unity Engine, where representations of the effects of visual obstacles to a human's perception can be simulated. The combined models will then determine the statistical probabilities of a Small Unmanned Aerial System (SUAS aka drone) becoming detected by enemy soldiers at various distances. The program will proceed to warn the drone operator when detection by an enemy soldier is likely.

In the future, we believe our work can be added by exploring the capabilities of cameras and sensors onboard actual SUAS platforms to dynamically locate targets and determine their distance, thereby directing itself to the best possible flight pattern to avoid detection.

View SUDO Poster

Smart Dashboard and Business Rule Engine

Faculty Advisor: Huiping Guo
Client: QTC
Liaisons: Francisco Guzman, Julian Gutierrez

Students: Alvent Chang, Jonathan Diaz, James Eddins, Bryan Gonzalez, Razin Khan, Pokuong Lao, Ashley Manese, Karina Pascual Zepeda, Adrian Salgado Lopez, Anthony Tsui

The purpose of the project is to develop a Smart Dashboard. The Smart Dashboard will consolidate system errors and user errors that have occurred in different systems. By using the best coding practices and an N-tier architecture, this Smart Dashboard will have the ability to support different lines of business, use dependency injection, support pagination, and display certain errors depending on what integration point the user clicked on.

(A) Smart Dashboard:

1. Users can fix integration issues related to data by knowing what the errors are. Developers can fix integration issues related to bugs in the system by knowing what the errors are.
2. This web app is coded in asp.net 6, bootstrap, javascript, and CSS and adheres to QTC's requirements and standards.
3. It currently supports the following integration points:

• Enterprise Printing Integration (Oracle)
• SQL

(B) Business Rule Engine:

The objective of the Business Rule Engine (BRE) is to receive a request with the name of a rule to execute that will yield a certain outcome by recursively making decisions. The advantage to such an engine is that there would no longer be a need to perform tasks such as classification manually; this is especially helpful if a user wants to classify a large dataset.

View Smartboard Poster  ViewBusiness Rule Poster

Southern California Edison VR Training Simulation

Faculty Advisor: David Krum
Client: Southern California Edison
Liaisons: Abder Elandaloussi

Students: Han Cao, Martin Castorena, Cameron Cheng, Manuel Guillen Vargas, Jaiden Holcomb, Seng Lei, Matthew Mendoza, Denise Tabilas, Fernando Torres, Kevin Truong

This senior design project aims to provide utility field workers with unique training opportunities through virtual reality simulation. The simulation is designed to enable field workers to gain practical experience in handling and installing equipment that they are expected to encounter in the field, thereby preparing them adequately for their actual field work. The project deliverables include a prototype using existing virtual reality devices available on the market today. The benefits from this project are situational awareness, field crew's safety, learning efficiency, structured guidance, and immediate auditory feedback as the simulation is carried out. The team worked with the project sponsors to investigate, learn, and understand the tasks of field workers at SCE who could benefit from using VR.

The team also critically analyzed VR technologies and platforms to find the best opportunities to improve safety, efficiency, and other benefits. After gathering project requirements, the team created a design alternative. The team gathered feedback from the sponsor and advisor throughout the project. After analyzing the design alternatives, the team created a unified system design, developed a prototype, and then evaluated the prototype. Documents, software code, and other deliverables will be shared with the sponsor.

View SCE Poster

The arqive

Faculty Advisor: John Hurley
Client: The arqive
Liaisons: Cynthia Wang, Zachary Vernon

Students: Cesar Ayala, Songtao Bu, Alejandro Ceballos, Bryan Chan, Erik Donovan-Blood, Kennard Lim, Jorge Mata Jr, Kalvin Mateo, Misael Ortega, Jonathan Saldivar, Dustin Shin

The arqive, formerly known as GlobaltraQs, is a web and mobile application that allows anyone to post fun and interesting LGBTQ+ oriented stories, events, and other information that they find meaningful—founded in 2014 by Dr. Cynthia Wang, with Zachary Vernon subsequently joining as a cosponsor. The arqive gives users a safe platform where they can share personal, historical, and community stories and access information about safe spaces, which all serve as valuable resources to members of the LGBTQ+ community. By allowing users to place pins on the map indicating where they have been and the experiences they have lived, The arQive allows people to mark their place in the world and in history.

This year's project for The arqive will focus on adding additional features to the website and mobile apps, creating an automated content moderation feature, adding further security measures for users who choose anonymity, adding a way to upload images and other media, beginning an AR/VR feature, and refining the appearance and performance of The arqive.

View Arqive Poster