SURP 2021 Project Abstracts
Summer Undergraduate Research Program
Faculty: Franz Kurfess
Email: email@example.comThe primary goal of this project is to explore the use of Artificial Intelligence and related computational methods in support of Search and Rescue missions. It is based on a collaboration with Gary Bloom, a Cal Poly CSC alumnus with a long involvement in S&R activities. As a first step, we propose to focus on the following aspects:
- Information Collection Script: Based on sources such as current practices, manuals, field guides, input from experienced personnel with different roles and perspectives, we will develop an adaptive script for S&R personnel to collect relevant information for the current mission. In a later stage, Natural Language Processing and Machine Learning methods can be used to generate the next set of questions on the fly based on the information provided so far.
- Information Consolidation: Relevant information from the first step is consolidated with the “field guide”, and presented to parties involved in the mission in an easily comprehensible and actionable manner. At the core will be a map with multiple layers of additional information such as roads or buildings that can be displayed as needed. In particular, we envision a “heat map” of a search area where hot spots indicate a high likelihood of finding a missing person. Since some of this processing may take place in areas with limited connectivity, we will design the system to provide the best available information under the circumstances.
- Operational Plan: Derive a plan for the operational activities of the current mission from the above information. This plan will include an assessment of priorities and probabilities for multiple scenarios, to be used by the mission management team. The team will have the ability to adjust input data and planning parameters. The planning scenario will be continually updated with information gathered from the ongoing S&R activities, e.g. to create the “next day” search plan.
- Wrap-Up and Assessment: Similar to the initial information collection script, we propose an Assessment Script with structured guidance to document a mission after the conclusion of the operational phase, This documentation can be extremely important in situations where criminal cases may be involved. It is also a valuable source of information for future missions, and such a template can be useful in systematically establishing a repository of missions.
- Analysis: The focus of a specific mission is on the immediate emergency at hand, such as finding a missing person. From a broader perspective, analyzing individual missions and sets of them can offer additional insights, such as the success and failure of strategies or specific actions, and recommendations for measures to prevent future emergencies. We propose the use of AI methods and tools such as Natural Language Processing, Knowledge Graphs, and Deep Learning to identify similarities across missions, compare outcomes with manuals and guidelines, or point out previously overlooked aspects.
Faculty: Mohamed Awwad
Email: firstname.lastname@example.orgRadio Frequency Identification (RFID) is a technology that enables the identification and tracking of tags attached to objects. It has been adopted over the past few decades by retailers and suppliers for better supply-chain-related decision-making. Presently, at Boeing, RFID data streams are primarily utilized for inventory management purposes, such as determining the accuracy of inventory count and deriving insight on material flow. This project aims to investigate ways of analyzing the immense volume of RFID data for items within Boeing’s facilities to enable better decision-making about the supply chain, specifically facility logistics. In addition, using the RFID data streams to produce insights such as the following will allow more efficient allocation of resources and optimize costs:
- Inventory flows and availability at different stages of the supply chain
- Frequent paths that products travel within facilities
- Identification of dead zones, zones with too much activity, and zones with too little activity
- Most active hours in a day, most active days in a month, or most active months in a year
Faculty: Trevor Cardinal
Our Laboratory in investigating skeletal muscle progenitor cells as a potential therapeutic candidate for peripheral artery disease. Maintaining the viability of transplanted cells requires the use of a transplant vehicle to provide the cells with a structural support. Students will evaluate cell distribution, viability, and phenotype in vitro and in vivo in two polymer hydrogels prepared by both bioprinting and manual extrusion. The specific hydrogels include a commercially-available formulation of pluronic-F127 with alginate and a customsynthesized tri-polymer prepared in the laboratory of Dr Philip Costanzo (Chemistry & Biochemistry) that has thermoresponsive properties (i.e. liquid at room temperature and semi-solid at body temperature) and is composed of poly(n-isopropylacrylamide), hydroxyethyl methacrylate, and hydroxyethyl arcylate.
Faculty: Joseph Callenes-Sloan
Affordable 3D printers, laser cutters, and Computer Numerical Control (CNC) mills will soon be available to millions of people. The technology has a wide reach, with the potential for broad social benefits. However, affordable 3D printers are often limited by accuracy, printing times, and reliability. This project proposes to address these limitations by adapting computer architecture techniques, which face similar problems, to the 3D printing system. All 3D printers today are built around an interface called G-code, which was first designed in G-code is a set of imperative commands that control the movement of the extruder and executes the printing. We will investigate the design of the 3D printing interface in the context of emerging 3D printing applications (e.g. biomedical applications where the compilation phase may take longer than the printing phase), and the dynamic characteristics of modern affordable 3D printing hardware. This project aims to improve the accuracy, performance, and reliability of 3D printing by adapting architectural techniques to address analogous printing problems.
Faculty: Maria Pantoja
As a part of the efforts by the California State University (CSU) System to achieve carbon neutrality by 2050, the Gold Tree Solar Farm (GTSF) was built on the San Luis Obispo campus. One of the challenges with integrating solar energy into the energy grid is the peak electrical demand does not coincide with peak power generation from solar fields. Solar energy is also considered an unreliable energy source because production varies significantly depending on the weather conditions. Utility providers are often required to balance solar generation to meet consumer demand, which often includes the costly process of activating/deactivating power plants. Therefore, there is considerable interest in increasing the accuracy and the granularity of solar power generation forecasting. This improved forecasting would reduce fluctuations of the electrical grid and facilitate its management. In this project, our goal is to be able to accurately forecast cloud coverage over a solar field up to 30 min into the future. The predictions are based on computer vision techniques which analyze total sky images deployed in the field.
Faculty: Mohsen Beyramali Kivy
In this proposed summer undergraduate educational research project, two undergraduate students will work with the PIs to create at least two practical design labs for the department’s freshman level courses, including MATE 110 and MATE 120 (Introduction to Materials Engineering Design I and II respectively). In these labs, freshman students will combine Theory, Granta Edupack Databases, and Experimentation to practice the concepts of Materials Engineering Design. The design labs will include project management, documentation in design, manufacturing techniques, and analysis of testing data. Created labs in this project will be implemented in Fall 2021 and Winter 2022 for the first time, the results and feedbacks will be collected and analyzed, and the necessary changes and improvements will be considered by the PIs for future years.
Faculty: Mohamed Awwad
Vaccination is considered the main breakthrough in the fight against the COVID-19 pandemic. With more people having access to the vaccine and hoping to get back to everyday life, there are still some doubts regarding equitable and fair access of some segments of the population to the COVID-19 vaccine. A part of this population that has not been prioritized yet to receive the COVID-19 vaccine is college students. However, and with ramping supply and availability of the vaccines over the next months, a plan must be developed to inoculate this age group with the vaccine. Allowing college-students to resume their normal daily activities, most importantly attending college in-person and in a safe environment.
This research project’s primary goal is to allow an interdisciplinary team of Cal Poly Engineering students the opportunity to develop a decision support system (DDS) to help a priority-based allocation and distribution of COVID-19 vaccines with a case applied to the Cal Poly campus. Additionally, and under the faculty advisor’s supervision, the students will test the developed DDS to enable campus leadership and other stakeholders to make real-time decisions regarding the distribution of COVID-19 among the student population. The DDS will integrate data analytics and simulation methods to solve vaccine allocation and optimal distribution problems.
Faculty: Jean Lee
The presence of microbial pathogens in water poses a high risk to public health worldwide. Current commonly used detection methods for these pathogens are time consuming and might not be accessible in certain parts of the world due to cost or resource limitations. This project aims to develop a nanotechnology-based method that can rapidly give a qualitative indication of the presence of water pathogens through color change, and quantitatively determine the concentrations of pathogens in water samples using non-sophisticated equipment. This detection technique has previously been reported in the literature using gold nanoparticles as the nanosensors. In this technique, the addition of the negatively-charged enzyme ß-galactosidase (ß-Gal), the dye chlorophenol red-β-D-galactopyranoside (CPRG), and positively-charged gold nanoparticles to a water sample without microbes will result in the ß-Gal binding to the gold nanoparticles via Coulombic attraction, and the color of the water will not change. However, when the negatively-charged microbes are present in the water sample, competitive binding will take place and the microbes will bind to the gold nanoparticles, freeing up the ß-Gal to react with the CPRG and change color. The magnitude of color change is directly proportional to the microbe concentration in the water sample. This project seeks to improve this method by using less expensive materials instead of gold for making the nanosensors, and investigating the re-use of nanosensor materials. Candidate sensor materials to be studied in this project include inexpensive insulator materials with low toxicity (such as sand or SiO2) and magnetic materials (such as iron oxide). These materials can be charged by rubbing with a wool or other type of cloth. After use, the sensor material is sterilized and re-used, and the performance of the candidate sensor materials when re-used will be studied. This project will also investigate whether positively-charged nanosensor materials can be eliminated altogether from this technique by using an electrically insulating material to contain the water sample (such as a clear glass vial) and positively charging the container. In this case, it is hypothesized that any negatively-charged microbes in the water sample will be attracted to the walls of the container, and that the ß-Gal and the CPRG that has been added to the water sample should react to change the color of the water to indicate the presence of microbes. If successful, the technology from this project can be commercialized and used for the testing of water quality in resource challenged locations to determine if their water is safe to drink and use.
Faculty: Jennifer Peuker
The project described here is a continuation of a curricular reimagining project designed to integrate social justice into a thermal systems design course and a machine design course, both taught within the Mechanical Engineering Department. There are myriad definitions of social justice, most of which support the idea that social justice is a value or belief that people should have equitable access to resources and protection of human rights. Engineers, whose impact on human lives is so far-reaching, have unique potential and unique responsibilities and accountability to society. To develop young engineers’ understanding of this potential and responsibility, discussions of social justice need to be integrated at multiple junctures through the engineering curriculum, especially in the spaces deemed purely technical.
To this end, through support from an Engineering IDEAS grant, Dr. Jennifer Mott Peuker worked with a team of students to implement three social justice modules into her thermal systems design course in Fall 2020. These modules were based on Dr. Donna Riley’s “Engineering Thermodynamics and 21stCentury Energy Problems: A Textbook Companion for Student Engagement”. Dr. Lauren Cooper used her IDEAS support to engage two students in the adaptation of two modules from Riley’s text to fit her mechanical systems design course context, which will be implemented in Spring and Summer 2021.
Our SURP students will analyze quantitative survey data and qualitative focus group data collected in Dr. Mott’s thermal systems design course. They will also help write an IRB proposal, generate surveys, and lead focus groups with Dr. Cooper’s Summer 2021 mechanical systems design course students. Lastly, we envision our students to actively help us refine and shape our social justice modules based on their own experience as students and what we collectively glean from preliminary data analysis.
Ideally, we would be able to meet and work in person during the summer, however, all tasks could be completed remotely as well. Because of this, students will not be required to physically be in San Luis Obispo for the summer to be able to work on this project.
Faculty: Long Wang
Electrical impedance tomography (EIT) technique is an electrical imaging method that determines the conductivity distribution of a conductive body. The EIT technique has attracted extensive attention due to its unique advantages in non-invasively mapping the spatial electrical properties of an object only using its boundary voltage measurements. The potential applications of this imaging technique include cancerous tumor detection, gastric function monitoring, non-destructive testing and evaluation of materials, among others. The typical mechanism of the EIT technique is that an electrical current is injected into the material of interest, and the induced boundary voltages are measured and used as inputs to EIT algorithms. In general, the EIT algorithm includes the forward and inverse problems. The forward problem aims to solve for boundary voltages based on an assumed conductivity distribution, whereas the inverse problem seeks to estimate the conductivity distribution of the material from boundary voltages. Built on our existing algorithms, this project will focus on using different computational techniques to effectively reconstruct the EIT inverse problem and achieve new functionalities including high-speed dynamic mapping and reconstruction uncertainty analysis. The performance of the developed EIT algorithms will be computationally evaluated based on simulated case studies. This research can be conducted virtually.
Faculty: John Bellardo
This project will advance Cal Poly’s next-generation avionics capabilities for small satellites. The current generation avionics, developed a decade ago, incorporated a then sector-leading combination of vertical integration, compactness, fault tolerance, software environment, and power consumption. Incremental improvements made in subsequent years have kept the platform relevant, but it is being surpassed by more recent clean-slate designs. As a result, Cal Poly is at a competitive disadvantage for the increasingly more sophisticated small satellite missions that we would like to develop. For instance, there are a number of opportunities to build and launch interplanetary spacecraft that are not well suited to the existing avionics. During the SURP the students will work with a prototype of the next-generation avionics in the CubeSat lab. They will focus on characterizing the prototype and providing recommendations for improvement. This process involves developing additional hardware and software necessary to effectively characterize performance, performing characterization, and making improvement recommendations for the next prototype based on how well the system meets mission requirements projected over the next decade.
Faculty: Alicia Johnstone
The project has two main objectives: 1) manufacture and verification of X-band antenna and transmitter and 2) development of software for UHF communication. One student will be allocated to each objective. The first objective is part of a technology demonstration project in which the overall goal is to design, manufacture, integrate and verify on orbit a capable X-band radio adapted for small satellites and their inherent constraints of available volume and funds. The use of such a band for communication is critical for the advancement of space sciences and exploration. Developing such a system available at a lower cost than the current commercially available solutions would enable expanded access to space, to a larger number of organizations within and outside the U.S. The second objective is part of the continuous improvement philosophy of the CubeSat Laboratory. The software to be developed will be based on the use of a new UHF chip and will enable improved communications using the amateur radio band. It will also contribute to the educational value of using of small satellites as a hands-on application for the learning of engineering concepts.
Faculty: Roger Benham
There is an increasing interest in the use of artificial reefs to address problems of sand and beach erosion and ecosystem enhancement. The reef experiment being proposed involves the design and testing of of a wave energy dissipation system, bio-habitat, and removable artificial reef concept. This project is a continuation of the 2019 SURP conducted by this sponsor. The 2019 SURP included materials and corrosion testing, with limited biological growth monitoring. The 2019 SURP project was met with great interest and enthusiasm by the students, and was a good learning internship experience in many ways, and culminated in a poster presentation during the 2019 Fall quarter. The team for the 2021 SURP would include two students, one from Biology and one from the Mechanical Engineering Department. The scope-of-work would include the installation of sample “coil” reef installed at the end of Cal Poly Pier (we already have an allotted space), monitoring for bio-growth on at least a weekly basis (may involve getting wet), and the design and building of a “drone” for monitoring the sample coil reef after the 2021 SURP has ended. This “drone” could be as simple as a camera on the end of a pvc pipe, or more complex (it is up to the students to decide what they can do on the time frame allowed). All expenses for the drone would be paid by Sponsor. A poster presentation of the results would be made during the Fall 2021 quarter.
Faculty: Pauline Faure
The project has two main objectives: 1) thermal analysis of a 3U spacecraft and 2) testing of maximum power point tracking based electrical power subsystem for the handling of 100W. One student will be allocated to each objective. This is part of the PowerSat project, a satellite under development at the Cal Poly CubeSat Laboratory. This is a technology demonstration project, which overall goal is to demonstrate the capabilities of a payload for the generation of a 100W of electrical power within a 1U volume. The CubeSat Laboratory is tasked to design, develop, integrate, and verify the satellite bus elements that support the payload functions. In January 2020, a proposal was submitted to NASA’s CubeSat Launch Initiative for a free launch opportunity during FY2023 and the achievement of the two objectives of the project is critical for the advancement PowerSat.
Faculty: Bruno da Silva
Software development is strongly dependent on source code management tools and platforms such as Git, GitHub, Bitbucket, and GitLab. Software teams also rely on issue tracking systems such as Jira, Bugzilla, and GitHub Issues, to manage tasks during the development and maintenance life-cycle of software applications. By using those systems, software developers leave a massive amount of valuable data regarding the product or service they develop and the process they follow to build and maintain these products/services – we call it software repository data. However, software repository data is only partially explored by services such as GitHub. When these data are utilized to provide useful information to developers, it is usually scattered over web-based interfaces in different tools. These tools and interfaces are not always convenient for developers when they have to make quick decisions during team meetings or when they are juggling through multiple screens that they have to use to carry out their engineering tasks. Moreover, it is common that software developers work in remote teams geographically distributed. This has become even more common after Covid- Therefore, in this project, we propose implementing a smart text-based assistant for software developers and software teams to execute on well-known text platforms for teams, such as Slack, Discord, and MS-Teams. This work builds on our previous experience in successfully developing a voice-assistant (last year’s SURP project and currently being extended as an MS thesis) for software developers. Our text-based assistant will gather data from software repository platforms such as GitHub and answer questions such as “Who was the last contributor to the module Customer.py?”, “Who has written more tests for Post.java?”, “Who has been assigned the highest number of issues in this sprint?”, “How many pending pull requests we have?”, “What’s our avg bug fixing time?”, “What’s our median pull request review time?”, “Who’s is assigned to issue #10?”, “What’s the build status of our main branch?” and some more complex questions, such as, “Who should review file App.js?”, “Who should be assigned to issue #15?”, “Have we increased test coverage in the last two sprints?”, “How much of our codebase has grown over time?”
Faculty: Franz Kurfess
The August 2020 CZU fire destroyed and damaged a significant number of structures at Cal Poly’s Swanton Pacific Ranch (SPR). This monumental loss, however, also offers new opportunities. The goal of this project is to explore opportunities for including “smart” technologies in the rebuilding, restoration, and re-vegetation efforts at the Ranch. This includes sensor networks, “smart building” technologies, and communication infrastructure on one hand, but also the use of AI methods such as Deep Learning and Knowledge Graphs for the above activities. This project will expand two related activities: Cal Poly’s Data Science Strategic Research Initiative, and an ad-hoc effort of a group of faculty and staff interested in data collection at SPR. It has the potential of involving students from a wide range of disciplines from all colleges at Cal Poly.
Faculty: Jacques Belanger
The 4.5 MW Gold Tree solar field on Cal Poly’s campus offers an invaluable opportunity for our students to work on a utility scale single-axis tracking solar field in collaboration with engineers from the company managing the field, REC Solar. The solar field was built on campus and the contract included access for Cal Poly researchers to real time data from the field as well as physical access to the site. Currently the energy generation of the solar field is below the original estimate due in large part to the skewed topography of the site. That topography has led to one row of panels shading adjacent rows in the morning and evening. Our research group, over the last two years, has developed a model of the solar field with the ultimate goal of optimizing the solar field energy generation. The model is at this point fairly accurate but we have recently discovered a few discrepancies that need to be further investigated. The students chosen for the project this summer would be investigating those discrepancies. The first one is related to the model predicting the power generated by a panel. To accurately predict that power you need to know the panel temperature and how it is affected by partial shading. The students will perform field measurements and shading experiments in our solar lab to test and validate the temperature and shading models used in our simulations. Another discrepancy that needs to be investigated is differences between data provided by the online data acquisition system operated by REC Solar and actual data measured in the field. An example is the tilt angle of the panels, where the observed tilt angle of the panel could be up to 4o off the value indicated by the data acquisition system. This angle difference changes the shading patterns on the panels and results in incorrect power generation prediction by the model. The students will compare their field measurements with values provided by the data acquisition system to identify discrepancies that are affecting the accuracy of the model. As a challenge, we will ask the students to design field measurement tools that could collect and store the data autonomously.
Faculty: Kristen Cardinal
The goal of Professor Kristen Cardinal’s research lab is to grow tissue engineered “blood vessel mimics” to use as early-stage living testing systems to evaluate vascular devices such as stents and flow diverters. These tissue engineered vessels consist of a polymer scaffold, or tube, lined with human endothelial cells, and cultivated in a bioreactor system. Over the past several years, the lab has established new techniques and implemented new components for creating customized and/or more optimized vessels. These recent advances have primarily been based upon new scaffold options, including more compliant electrospun materials and more customizable silicone geometries. As the lab continues to expand and build upon these areas, it is important to understand the reproducibility and scalability of the new techniques. The goal of this SURP project is to have a team of 2 students working together on scaffold fabrication and vessel cultivation using these new techniques in order to characterize reproducibility. One student will specifically focus on the scaffold fabrication aspect, learning and performing electrospinning, while the other student will learn how to create vessels to carry out the vessel cultivation aspect. In order to be ready to evaluate reproducibility over summer, each student will need to receive preliminary training in their respective techniques prior to June 2021.