2022 Research Projects
Projects are posted below; new projects will continue to be posted. To learn more about the type of research conducted by undergraduates, view the archived symposium booklets and search the past SURF projects.
This is a list of research projects that may have opportunities for undergraduate students. Please note that it is not a complete list of every SURF project. Undergraduates will discover other projects when talking directly to Purdue faculty.
You can browse all the projects on the list or view only projects in the following categories:
Big Data/Machine Learning (45)
A Hyperspectral imager for Propulsion Testing
Description:
The SURF student will work with a PhD student and Professor Gore to contribute to the following project.
The feasibility of utilizing a mid-infrared hyperspectral imager as a general-purpose ground testing diagnostic for rocket propulsion systems will be demonstrated. Purdue University compared temperatures deconvoluted from the hyperspectral images of a hydrogen air premixed flame with our measurements of temperature using Rayleigh scattering at identical operating conditions. This comparison has helped sponsor address a key issue involving the establishment of feasibility of obtaining spatially and temporally resolved information from mid-infrared hyperspectral imager measurements.
Purdue University work in Phase II is focused on evaluating the imager for different flame configurations to help deliver the system to NASA. Purdue University will demonstrate the use of the hyperspectral imager using: (i) two previously studied turbulent premixed hydrogen air jet flames, (ii) two previously studied turbulent premixed and partially premixed methane air jet flames, and (iii) a plume emerging from an existing rocket propellant combustion test apparatus.
The feasibility of utilizing a mid-infrared hyperspectral imager as a general-purpose ground testing diagnostic for rocket propulsion systems will be demonstrated. Purdue University compared temperatures deconvoluted from the hyperspectral images of a hydrogen air premixed flame with our measurements of temperature using Rayleigh scattering at identical operating conditions. This comparison has helped sponsor address a key issue involving the establishment of feasibility of obtaining spatially and temporally resolved information from mid-infrared hyperspectral imager measurements.
Purdue University work in Phase II is focused on evaluating the imager for different flame configurations to help deliver the system to NASA. Purdue University will demonstrate the use of the hyperspectral imager using: (i) two previously studied turbulent premixed hydrogen air jet flames, (ii) two previously studied turbulent premixed and partially premixed methane air jet flames, and (iii) a plume emerging from an existing rocket propellant combustion test apparatus.
Research categories:
Big Data/Machine Learning, Energy and Environment, Thermal Technology
Preferred major(s):
- Aeronautical and Astronautical Engineering
- Mechanical Engineering
School/Dept.:
Mechanical Engineering
Professor:
Jay
Gore
More information: https://engineering.purdue.edu/GRG
AAMP UP- Computational Study of Energetic Materials
Description:
This project is part of the AAMP-UP '22 program, which focuses on energetic material research.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Marcial Gonzalez and his team. The goal is to enable discovery of the physical mechanisms driving impact-induced reactions in explosives (such as interparticle relative velocities and contact pressures) and their correlation with material properties, formulation, and processing. Description: Energetic materials used in the propulsion industry, and particle-binder composites in general, consist of (explosive) crystals in a binder matrix. The mechanical properties and performance of these materials are strongly correlated to their microstructure. Specifically, the mechanical response under weak impact loads is characterized by long-range microstructural correlations (typical of granular materials) that need to be resolved with high statistical significance in order to understand how complex phenomena (such as detonation) emerges from local simpler interactions. The interplay between these long-range interactions and local mechanisms (such as deformation, failure and, ultimately, hot spot formation) is responsible for the sensitivity of the material. A mesoscale modeling methodology, capable of modeling the dynamic response of large energetic composites (>50,000 crystals) by analytically upscaling complex crystal-binder deformation and failure mechanics into closed-form contact laws, will be used to build time-dependent multivariate probability distribution functions of microstructural features.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Marcial Gonzalez and his team. The goal is to enable discovery of the physical mechanisms driving impact-induced reactions in explosives (such as interparticle relative velocities and contact pressures) and their correlation with material properties, formulation, and processing. Description: Energetic materials used in the propulsion industry, and particle-binder composites in general, consist of (explosive) crystals in a binder matrix. The mechanical properties and performance of these materials are strongly correlated to their microstructure. Specifically, the mechanical response under weak impact loads is characterized by long-range microstructural correlations (typical of granular materials) that need to be resolved with high statistical significance in order to understand how complex phenomena (such as detonation) emerges from local simpler interactions. The interplay between these long-range interactions and local mechanisms (such as deformation, failure and, ultimately, hot spot formation) is responsible for the sensitivity of the material. A mesoscale modeling methodology, capable of modeling the dynamic response of large energetic composites (>50,000 crystals) by analytically upscaling complex crystal-binder deformation and failure mechanics into closed-form contact laws, will be used to build time-dependent multivariate probability distribution functions of microstructural features.
Research categories:
Big Data/Machine Learning, Deep Learning
Preferred major(s):
- No Major Restriction
Desired experience:
MATLAB knowledge.
Must be a U.S. citizen, national, or permanent resident of the United States. Must have completed at least one academic semester of full-time study at associate's or bachelor's degree level from an accredited college or university.
School/Dept.:
Mechanical Engineering
Professor:
Marcial
Gonzalez
More information: https://engineering.purdue.edu/ME/People/ptProfile?resource_id=106137
AAMP UP- Machine Learning Applied to Explosives
Description:
This project is part of the AAMP-UP '22 program, which focuses on energetic material research.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Steven Son and his team. Machine learning (ML) tools are playing an increasingly important role in science and engineering, revealing patterns and providing predictive capabilities not achievable otherwise. This research area explores the utility of machine learning algorithms in the design, development, and characterization of various energetic material systems. Particular emphasis is placed on bringing a data science formalism to the field, with an eye toward both future capability development and more intelligent (and appreciably faster) material formulation and system design. The REU student would work closely with a Research Scientist and graduate student to gather data, analyze it using ML tools, and share these results.
Some experience w/ coding, AI, or ML recommended.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Steven Son and his team. Machine learning (ML) tools are playing an increasingly important role in science and engineering, revealing patterns and providing predictive capabilities not achievable otherwise. This research area explores the utility of machine learning algorithms in the design, development, and characterization of various energetic material systems. Particular emphasis is placed on bringing a data science formalism to the field, with an eye toward both future capability development and more intelligent (and appreciably faster) material formulation and system design. The REU student would work closely with a Research Scientist and graduate student to gather data, analyze it using ML tools, and share these results.
Some experience w/ coding, AI, or ML recommended.
Research categories:
Big Data/Machine Learning, Deep Learning, Material Modeling and Simulation
Preferred major(s):
- No Major Restriction
Desired experience:
Some experience w/ coding, AI, or ML recommended.
Must be a U.S. citizen, national, or permanent resident of the United States. Must have completed at least one academic semester of full-time study at associate's or bachelor's degree level from an accredited college or university.
School/Dept.:
Mechanical Engineering
Professor:
Steven
Son
More information: https://engineering.purdue.edu/ME/People/ptProfile?resource_id=29385
AAMP UP- Machine Learning Approaches to Energetic Materials
Description:
This project is part of the AAMP-UP '22 program, which focuses on energetic material research.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Alejandro Strachan and his team. The project will use tools from data science and machine learning to develop predictive models for the performance of energetic materials. Students will learn about neural networks, deep learning, and the chemistry and physics of energetic materials.
Basic python knowledge is desirable for the project.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Alejandro Strachan and his team. The project will use tools from data science and machine learning to develop predictive models for the performance of energetic materials. Students will learn about neural networks, deep learning, and the chemistry and physics of energetic materials.
Basic python knowledge is desirable for the project.
Research categories:
Big Data/Machine Learning, Deep Learning
Preferred major(s):
- No Major Restriction
Desired experience:
Basic python knowledge.
Must be a U.S. citizen, national, or permanent resident of the United States. Must have completed at least one academic semester of full-time study at associate's or bachelor's degree level from an accredited college or university.
School/Dept.:
Materials Engineering
Professor:
Alejandro
Strachan
More information: https://engineering.purdue.edu/MSE/people/ptProfile?id=33239
AAMP UP-Machine Learning Based Development of Multiscale Reactive Model of High Explosives
Description:
This project is part of the AAMP-UP '22 program, which focuses on energetic material research.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Alejandro Strachan and his team. Computationally efficient models for accurately predicting the thermal and chemical properties of high explosive (HE) materials are of great interest for both military and civilian applications. In this project, students will work on developing a multiscale model for HE materials, namely HMX and TATB, via combination of dimensionality reduction through machine learning techniques and atomistic molecular dynamics (MD) simulations. Specifically, the decomposition behavior of HE will be investigated through reactive, atomistic MD simulations. The dimensionality of the thermal and chemical data collected from the MD will then be reduced through non-negative matrix factorization technique to develop reduced-order chemistry models for HMX and TATB. Utilizing these, continuum-level models that capture the decomposition behavior of these materials will be developed.
Students with programming experience would be a good fit for this project.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Alejandro Strachan and his team. Computationally efficient models for accurately predicting the thermal and chemical properties of high explosive (HE) materials are of great interest for both military and civilian applications. In this project, students will work on developing a multiscale model for HE materials, namely HMX and TATB, via combination of dimensionality reduction through machine learning techniques and atomistic molecular dynamics (MD) simulations. Specifically, the decomposition behavior of HE will be investigated through reactive, atomistic MD simulations. The dimensionality of the thermal and chemical data collected from the MD will then be reduced through non-negative matrix factorization technique to develop reduced-order chemistry models for HMX and TATB. Utilizing these, continuum-level models that capture the decomposition behavior of these materials will be developed.
Students with programming experience would be a good fit for this project.
Research categories:
Big Data/Machine Learning, Deep Learning
Preferred major(s):
- No Major Restriction
Desired experience:
Programming experience a plus
Must be a U.S. citizen, national, or permanent resident of the United States. Must have completed at least one academic semester of full-time study at associate's or bachelor's degree level from an accredited college or university.
School/Dept.:
Materials Engineering
Professor:
Alejandro
Strachan
More information: https://engineering.purdue.edu/MSE/people/ptProfile?id=33239
AI in manufacturing and deployment of TinyML devices.
Description:
Private project for Eugene Thomson
Research categories:
Big Data/Machine Learning
Preferred major(s):
- No Major Restriction
School/Dept.:
ECE
Professor:
Ali
Shakouri
Advanced Vehicle Automation and Human-Subject Experimentation
Description:
Vehicle automation is developing at a rapid rate worldwide. While fully autonomous vehicles will be pervasive on the roadway for the next several years, many research initiatives are currently underway to understand and design approaches that will make this technology a future reality. This work ranges from the development of sensors and controls algorithms, to schemes for networks and connectivity, to the creation of in-vehicle driver interfaces. Here, one component that is key to the effective design of next-generation autonomous driving systems is the human driver and, thus studying human-vehicle interactions and defining driver’s roles/tasks will be important.
The goal of this project is to describe and measure the ways in which a person interacts with advanced vehicle automation. Students will assist with multiple activities and will learn a combination of the following: how to a) develop/code advanced driving simulation scenarios, b) collect driving performance data, c) analyze driver and performance data (using methods via software packages), and d) write technical reports and/or publications. Students may also gain experience collecting and analyzing complementary physiological measures, such as eye movement data, brain activity, skin conductance, and heart rate. The students will work closely with graduate student mentors to enhance learning.
The goal of this project is to describe and measure the ways in which a person interacts with advanced vehicle automation. Students will assist with multiple activities and will learn a combination of the following: how to a) develop/code advanced driving simulation scenarios, b) collect driving performance data, c) analyze driver and performance data (using methods via software packages), and d) write technical reports and/or publications. Students may also gain experience collecting and analyzing complementary physiological measures, such as eye movement data, brain activity, skin conductance, and heart rate. The students will work closely with graduate student mentors to enhance learning.
Research categories:
Big Data/Machine Learning, Learning and Evaluation, Other
Preferred major(s):
- No Major Restriction
Desired experience:
Human Factors, Matlab, transportation, some experience in statistics, some computer programming and machine learning experience (in any language)
School/Dept.:
Industrial Engineering
Professor:
Brandon
Pitts
More information: https://engineering.purdue.edu/NHanCE
Advancing Pharmaceutical Manufacturing through Process Modeling and Novel Sensor Development
Description:
The limitations of batch processes to manufacture pharmaceutical products such as tablets, coupled with advances in process analytical technology (PAT) tools have led to a shift towards continuous manufacturing (CM), which represents the future of the pharmaceutical industry.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
Research categories:
Big Data/Machine Learning, Chemical Unit Operations, Material Processing and Characterization
Preferred major(s):
- No Major Restriction
Desired experience:
Basic skills for MATLAB and powder characterization would be a plus, but they are not necessary. The student should be safety conscious, self-motivated, and can work with minimal supervision. Aptitude for mastering the use of gadgets is desired, as well as the ability to understand research papers, documents, and manuals.
Any student who prefers a combination of simulation/modeling and hands-on pilot plant work is welcome. Moreover, this project is ideal for a student who is interested in a career in pharma or in powder manufacturing.
School/Dept.:
Davidson School of Chemical Engineering
Professor:
Gintaras
Reklaitis
Advancing Pharmaceutical Manufacturing through Process Modeling and Novel Sensor Development
Description:
The limitations of batch processes to manufacture pharmaceutical products such as tablets, coupled with advances in process analytical technology (PAT) tools have led to a shift towards continuous manufacturing (CM), which represents the future of the pharmaceutical industry.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
Research categories:
Big Data/Machine Learning, Chemical Unit Operations, Material Processing and Characterization
Preferred major(s):
- No Major Restriction
Desired experience:
Basic skills for MATLAB and powder characterization would be a plus, but they are not necessary. The student should be safety conscious, self-motivated, and can work with minimal supervision. Aptitude for mastering the use of gadgets is desired, as well as the ability to understand research papers, documents, and manuals.
Any student who prefers a combination of simulation/modeling and hands-on pilot plant work is welcome. Moreover, this project is ideal for a student who is interested in a career in pharma or in powder manufacturing.
School/Dept.:
Davidson School of Chemical Engineering
Professor:
Gintaras
Reklaitis
Advancing Pharmaceutical Manufacturing through Process Modeling and Novel Sensor Development
Description:
The limitations of batch processes to manufacture pharmaceutical products such as tablets, coupled with advances in process analytical technology (PAT) tools have led to a shift towards continuous manufacturing (CM), which represents the future of the pharmaceutical industry.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
The flexibility of continuous processes can reduce wasted materials and facilitate scale-up more easily with active plant-wide control strategies. Ultimately, this results in cheaper and safer drugs, as well as a more reliable drug supply chain.
To fully realize the benefits of continuous manufacturing, it is important to capture the dynamics of the particulate process, which can be more complex than common liquid-based or gas-based chemical processes. In addition, effective fault detection and diagnostic systems need to be in place, so intervention strategies can be implemented in case the system goes awry.
All of these require the development of process models that leverages knowledge of the process and big data. Students in this part of the research would have a chance to gain experience in industry-leading software for process modeling (e.g. Simulink, gProms, OSI PI) and machine learning (e.g. Matlab, Python, .NET).
Most importantly, they would be able to test the models in Purdue's Newly Installed Tablet Manufacturing Pilot Plant at the FLEX Lab in Discovery Park.
Another important aspect of the research are sensors. In this project, we will be investigating the feasibility of two novel sensors: a capacitance-based sensor to measure mass flow, and a particle imaging sensor that directly captures images of the powder particles to give you a particle size distribution. We will be testing these sensors together with NIR and Raman sensors, and use data analytics to determine their feasibility of application in a drug product manufacturing process.
Research categories:
Big Data/Machine Learning, Chemical Unit Operations, Material Processing and Characterization
Desired experience:
Basic skills for MATLAB and powder characterization would be a plus, but they are not necessary. The student should be safety conscious, self-motivated, and can work with minimal supervision. Aptitude for mastering the use of gadgets is desired, as well as the ability to understand research papers, documents, and manuals.
Any student who prefers a combination of simulation/modeling and hands-on pilot plant work is welcome. Moreover, this project is ideal for a student who is interested in a career in pharma or in powder manufacturing.
School/Dept.:
Davidson School of Chemical Engineering
Professor:
Gintaras
Reklaitis
Anomaly Detection in Extrusion Based Additive Manufacturing
Description:
This project is a part of Prof. Monique McClain's team. Cheap commercial Fused Filament Fabrication 3D printers do not currently have the ability to sense when a print error is occurring (e.g. delimination, overextrusion, etc.). If an operator is not monitoring the print or has limited experience, then this will lead to the creation of defective parts and wasted material, time, and effort. In order to move towards automatic on-the-fly correction during a print, it is important to be able to measure and classify critical 3D printing defects. By using an existing 3D printer with integrated sensors, the research assistant will have to design and conduct printing experiments that can allow us to distinguish normal (good) operation from abnormal (bad) operation. Then, machine learning/statistical process control algorithms can be applied to the carefully collected data in order to test the accuracy of the algorithms in detecting such errors. Ultimately, the research assistant will gain experience in additive manufacturing, design of experiments, and data analysis from this project.
Research categories:
Big Data/Machine Learning, Other
Preferred major(s):
- No Major Restriction
Desired experience:
Python, 3D printing, linear algebra, sensor measurement, statistics
School/Dept.:
Mechanical Engineering
Professor:
Monique
McClain
Data Driven Modeling of Electric Vehicle Impacts on Traffic Safety
Description:
As the Biden administration recently announced a national target for electric vehicle (EV) sales, more and more EVs will be on road in the future. Meanwhile, there will be an increasing possibility of traffic crashes between EVs and Internal Combustion Engine (ICE) vehicles or between EVs and pedestrians/cyclists. However, we have a limited understanding of how EVs will influence traffic safety, especially at road intersections. This study will leverage affluent historical traffic crash data (including driver demographic information, driver behavior, and traffic conditions) in Indiana and conduct data-driven modeling to uncover what factors are associated with crashes involving EVs. In specific, this study will focus on crashes on all interstate and state highways in Indiana. The expected outcome will lead to policy recommendations on developing EV safety regulations, improving vehicle safety features and highway design in the future.
Research categories:
Big Data/Machine Learning, Energy and Environment, Other
Preferred major(s):
- Civil Engineering
- Computer Science
- Statistics - Applied Statistics
Desired experience:
This research will involve statistical modeling and spatial-temporal data analysis and require basic programming skills (e.g., Python or R). Other desired qualifications include ability to work independently, strong work ethic, ability to work in diverse teams, and tehnical writing skills.
School/Dept.:
Lyles School of Civil Engineering
Professor:
Konstantina (Nadia)
Gkritza
More information: https://engineering.purdue.edu/ASPIRE; https://engineering.purdue.edu/STSRG
Data-based Explanations for Fairness Debugging of Decision Trees
Description:
About the project: Machine learning (ML) models are increasingly being used in a wide range of domains such as loan applications, healthcare, crime prevention and justice management. However, there is mounting concern that the complexity and opacity of ML models perpetuates systemic biases and discrimination reflected in training data. This project centers at the (realistic) assumption that the bias in the predictions of an ML model is, in large part, due to the data used to train the model. We are interested in identifying coherent parts of the training data responsible for the bias. Such explanations are termed as data-based explanations. We have recently developed a system that generates interpretable data-based explanations for ML models with (twice-differentiable) loss functions. In the proposed SURF project, we are interested in generating data-based explanations for tree-based ML models such as decision trees and gradient boosted decision trees.
Importance of the project: This project is a part of our efforts to generate explanations for ML model outcomes and bias. Results from the project will be instrumental in advancing the state of the art in the area of explainable artificial intelligence.
Scope of the project, and the student's contribution: The project will focus at a specfic class of ML models (decision trees). Given an ML model and a dataset used to train the model, we will develop efficient algorithms to identify subsets of the training data that are the most responsible for the bias of the model. The student is expected to have basic knowledge of machine learning (in particular, supervised learning and decision trees), and will participate in developing and implementing the solutions in Python.
Importance of the project: This project is a part of our efforts to generate explanations for ML model outcomes and bias. Results from the project will be instrumental in advancing the state of the art in the area of explainable artificial intelligence.
Scope of the project, and the student's contribution: The project will focus at a specfic class of ML models (decision trees). Given an ML model and a dataset used to train the model, we will develop efficient algorithms to identify subsets of the training data that are the most responsible for the bias of the model. The student is expected to have basic knowledge of machine learning (in particular, supervised learning and decision trees), and will participate in developing and implementing the solutions in Python.
Research categories:
Big Data/Machine Learning
Preferred major(s):
- Computer Science
- Computer Engineering
- Computer and Information Technology
Desired experience:
Introductory course in Machine Learning or Data Science
School/Dept.:
Computer and Information Technology
Professor:
Romila
Pradhan
Distributed Deep Learning for Multi-Robot Control
Description:
This project aims to design new deep learning/machine learning approaches to control two robots navigating through an environment. The robots aim to achieve a certain destination while avoiding obstacles and performing specific tasks, e.g., making a delivery along the way. The robots further perform information sharing and collaborate with each other to further increase their capability (the robot info:https://www.robotis.us/turtlebot-3-waffle-pi/). This setup aims to provide proof-of-concept for the efficacy of distributed deep learning models in a realistic multi-robot navigation task. The role of the undergraduate research would be to contribute to algorithm design, implementation, and benchmarking steps.
Research categories:
Big Data/Machine Learning, Deep Learning
Preferred major(s):
- No Major Restriction
School/Dept.:
ECE
Professor:
Abolfazl
Hashemi
More information: https://abolfazlh.github.io/
Energetic Particle Adhesion via enhanced centrifuge method
Description:
This project is part of the AAMP-UP '22 program, which focuses on energetic material research.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Stephen Beaudoin and his team. The research will use tools from data science and machine learning to develop predictive models for the performance of energetic materials. Students will learn about neural networks, deep learning, and the chemistry and physics of energetic materials.
AAMP-UP is separate but highly partnered with SURF.
The project is run by Dr. Stephen Beaudoin and his team. The research will use tools from data science and machine learning to develop predictive models for the performance of energetic materials. Students will learn about neural networks, deep learning, and the chemistry and physics of energetic materials.
Research categories:
Big Data/Machine Learning, Deep Learning, Other
Preferred major(s):
- No Major Restriction
Desired experience:
Must be a U.S. citizen, national, or permanent resident of the United States. Must have completed at least one academic semester of full-time study at associate's or bachelor's degree level from an accredited college or university.
School/Dept.:
Chemical Engineering
Professor:
Stephen
Beaudoin
More information: https://engineering.purdue.edu/ChE/people/ptProfile?resource_id=11574
Forecasting the 2022 U.S. Elections using Mathematical Modeling
Description:
Election forecasting involves polling likely voters, weighting polling data, combining it with other information (e.g., how the economy is doing), accounting for uncertainty, and communicating forecasts to the public. In this project, we will use mathematical modeling to produce forecasts of the upcoming 2022 U.S. midterm elections, and we will build a website to post our election forecasts.
Research categories:
Big Data/Machine Learning, Other
Preferred major(s):
- No Major Restriction
- Computer Science
- Mathematics
- Civil Engineering
- Biomedical Engineering
Desired experience:
Good team members with experience in linear algebra and differential equations, interest in interdisciplinary research, and strong programming skills (Matlab, Html/Css, preferred but not necessary).
School/Dept.:
Mathematics
Professor:
Alexandria
Volkening
More information: https://modelingelectiondynamics.gitlab.io/2020-forecasts/index.html
Functional Near-InfraRed Spectroscopy (fNIRS) Testbed Development for Studying Human Interaction with Autonomous Systems
Description:
Bio or physiological sensors have become ubiquitous, and this additional sensing capability is being integrated into clothing and other devices (e.g. smart watches) that we wear to constantly provide us with feedback about our health. Physiological sensors can also be used to infer mental processes occurring in the brain; this is called "psychophysiological" sensing. In our lab we are studying the use of various types of psychophysiological measurements to infer or predict human decision-making and other cognitive factors (such as their trust) during interactions with machines, robots, and other autonomous systems. This type of research is important for improving the safety and performance of autonomous systems designed to interact with or assist humans, such as autonomous vehicles, nurse robots, or surgical robots used by physicians. We are looking for a student to set up a new sensor, called functional Near-InfraRed Spectroscopy (fNIRS), for use in our lab and to test it against some prior work we've done with similar sensors to study human trust in automation.
Research categories:
Big Data/Machine Learning, Human Factors, Other
Preferred major(s):
- Mechanical Engineering
- Industrial Engineering
- Electrical Engineering
- Aeronautical and Astronautical Engineering
- Civil Engineering
Desired experience:
Prior experience or familiarity with sensors, hardware, and interfacing such devices with software is desirable.
School/Dept.:
School of Mechanical Engineering
Professor:
Neera
Jain
More information: https://engineering.purdue.edu/JainResearchLab/
Geospatially resolved model of heat pump operating costs & emissions
Description:
Commercial and residential buildings account for 13% of greenhouse gas emissions in the United States. Most of these emissions are driven by heating and cooling energy demand. Heat pumps, especially new technologies that make use of low global warming potential refrigerants, offer higher energy efficiency but can still be cost prohibitive. Working as part of a larger research project, this summer project would use performance testing data of a new heat pump technology to estimate the energy consumption for operating a new heat pump technology at different locations with different weather conditions in the United States. Using this energy consumption data, we can then estimate the greenhouse gas emissions associated with these new technologies using today’s electricity grid, and the cost of operating these new heat pumps. The results will be incorporated into a broader study of the geospatial environmental and energy burdens to residents.
Research categories:
Big Data/Machine Learning, Energy and Environment
Preferred major(s):
- Mechanical Engineering
- Industrial Engineering
- Chemical Engineering
- Environmental and Ecological Engineering
- Electrical Engineering
Desired experience:
Completed introductory thermodynamics and electrical engineering coursework (ME200, ECE20001 or similar). Working familiarity with Python and/or Matlab. Experience with data science and/or GitHub is also a plus.
School/Dept.:
Mechanical Engineering
Professor:
Rebecca
Ciez
High speed 3D microscopy imaging
Description:
This project is NSF REU project that aims to develop a high-speed 3D microscopy imaging system for robotics, biological applications. Undergraduate will work with a graduate student to develop novel image processing algorithms , data analytics methods, and instrumentation.
Research categories:
Big Data/Machine Learning, Biological Characterization and Imaging, Fabrication and Robotics
Preferred major(s):
- No Major Restriction
Desired experience:
Prior programming experiences.
School/Dept.:
Mechanical Engineering
Professor:
Song
Zhang
More information: http://www.xyztlab.com
How do zebrafish get their stripes A mathematical and computational study
Description:
From leopards to fish, many animals sport patterns (like stripes or spots) on their bodies. My group takes a mathematical approach to understand how patterns form in the skin of zebrafish, which are small striped with important biomedical applications. Zebrafish development takes months, but simulating pattern formation takes minutes. In this project, I will mentor a student in building image-processing software to make simulated zebrafish patterns look more like real fish.
Research categories:
Big Data/Machine Learning, Biological Characterization and Imaging, Biological Simulation and Technology, Cellular Biology
Preferred major(s):
- Mathematics
- Computer Science
- Biomedical Engineering
- Biological Engineering - multiple concentrations
- Engineering (First Year)
- Agricultural Engineering
Desired experience:
Good team members who are excited about interdisciplinary research, have taken a course in linear algebra, and have strong programming skills.
School/Dept.:
Mathematics
Professor:
Alexandria
Volkening
More information: https://www.alexandriavolkening.com/agentBased.html
Human Factors: Enhancing Performance of Nurses and Surgeons
Description:
High physical and cognitive workload among surgeons and nurses are becoming more common. The purpose of this project is to examine the contributors to these and develop technology to understand and enhance their performance.
The SURF student will participate in data collection in the operating room at Indiana University School of Medicine, data analysis and interpretation, and write his/her results for a journal publication. The student will regularly communicate his/her progress and results with faculty, graduate mentors, and surgeon collaborators.
More information: https://engineering.purdue.edu/YuGroup
The SURF student will participate in data collection in the operating room at Indiana University School of Medicine, data analysis and interpretation, and write his/her results for a journal publication. The student will regularly communicate his/her progress and results with faculty, graduate mentors, and surgeon collaborators.
More information: https://engineering.purdue.edu/YuGroup
Research categories:
Big Data/Machine Learning, Human Factors, Learning and Evaluation, Medical Science and Technology
Preferred major(s):
- No Major Restriction
- Industrial Engineering
- Computer Science
- Biomedical Engineering
Desired experience:
Human Factors, Machine Learning, Sensors, Programming
School/Dept.:
Industrial Engineering
Professor:
Denny
Yu
More information: https://engineering.purdue.edu/YuGroup
Illumination of Damage through X-ray analysis
Description:
Damage in structural materials is often difficult to quantify, instead we rely on large scale component level testing and curve fitting. With the advent of advanced high energy X-ray characterization tools, including diffraction and tomography, we have the ability to identify damage inside the bulk of the material, in which the samples are subjected to mechanical loading. Thus, in this project, X-ray data will be reconstructed and the damage will be characterized and quantified in several material systems (including carbon fiber reinforced composites and Ti-6Al-4V produced via additive manufacturing). The interaction of damage with microstructural features will be assessed, in order to achieve a physics-based understanding of material failure.
Research categories:
Big Data/Machine Learning, Composite Materials and Alloys, Material Modeling and Simulation, Material Processing and Characterization
Preferred major(s):
- Aeronautical and Astronautical Engineering
- Materials Engineering
- Mechanical Engineering
- Computer Science
- Computer Engineering
Desired experience:
Students are expected to work with Image Processing and Visualization tools, as well as Matlab or Python.
School/Dept.:
School of Aeronautics and Astronautics
Professor:
Michael
Sangid
More information: https://engineering.purdue.edu/~msangid/
Industrial IoT Implementation and Machine Learning for Smart Manufacturing
Description:
The student will work with PhD students on implementation of IoT technology on manufacturing machines and processes, database development, dashboard development, and machine learning for smart manufacturing.
Research categories:
Big Data/Machine Learning, Deep Learning, Fabrication and Robotics, Internet of Things
Preferred major(s):
- Mechanical Engineering
- Computer Engineering
- Computer Science
School/Dept.:
Mechanical Engineering
Professor:
Martin
Jun
More information: https://web.ics.purdue.edu/~jun25/
Leveraging the Power of Social Networks to Eradicate Epidemics
Description:
Since the popularization of handheld communication devices and social media applications, opinion dynamics and social networks have played a more critical role in politics, economics, and public health issues. In particular, opinion polarization on vaccination has tolled thousands of lives in the recent pandemic. Consider the following question: "If you could only convince three nodes in a social network to get vaccinated, which nodes should you choose?"
This project will guide students to answer this resource allocation problem through analyzing the spread transmission network and the dynamic opinion network. The project will be composed of four parts:
1. Constructing epidemic spread simulators.
2. Designing a control strategy for epidemic mitigation.
3. Developing mathematical proofs which guarantee the algorithm's performance.
4. Applying the strategy to real networks generated from online COVID data as a case study.
Students who participated in the project will learn the basics of the epidemic modeling paradigm, network science, control theory, and Python/MATLAB programming skills.
This project will guide students to answer this resource allocation problem through analyzing the spread transmission network and the dynamic opinion network. The project will be composed of four parts:
1. Constructing epidemic spread simulators.
2. Designing a control strategy for epidemic mitigation.
3. Developing mathematical proofs which guarantee the algorithm's performance.
4. Applying the strategy to real networks generated from online COVID data as a case study.
Students who participated in the project will learn the basics of the epidemic modeling paradigm, network science, control theory, and Python/MATLAB programming skills.
Research categories:
Big Data/Machine Learning, Engineering the Built Environment, Learning and Evaluation, Medical Science and Technology
Preferred major(s):
- No Major Restriction
School/Dept.:
Elmore Family School of Electrical Engineering
Professor:
Philip E.
Paré
Machine learning-based modeling of linear and non-linear deformation in high-pressure hydrostatic machines
Description:
Machine learning, image processing, fluid-structure interaction, linear and nonlinear deformation, elasto-hydrodynamic lubrication... You may have learned or have heard some of those topics. But you may never see how those interdisciplinary techniques can be used together to solve a real-life engineering problem. This project offers you a unique experience to participate in my research group developing a first-in-kind machine learning-based simulation model for nonlinear contact problems in high-pressure hydrostatic machines. The objective of the SURF project is to create a machine learning algorithm capable of fast predicting the two-dimensional, nonlinear deformation distribution from the pressure distribution. The project will be rewarding and challenging, and your work will constitute a new modeling approach in the fluid power field. Therefore, it is also very possible to be published with you as a co-author.
You will be challenged to 1) learn to program a machine learning algorithm in TensorFlow, 2) generate a training dataset for machine learning models using a state-of-the-art numerical simulation tool, and 3) integrate the neural network into the existing modeling suite.
You will be supported by your graduate mentor, who specializes in these topic areas and will provide guidance throughout the project. You will also be supported by a group of 8 developers of the hydrostatic machine modeling toolset that are working on different aspects of the code.
You will be challenged to 1) learn to program a machine learning algorithm in TensorFlow, 2) generate a training dataset for machine learning models using a state-of-the-art numerical simulation tool, and 3) integrate the neural network into the existing modeling suite.
You will be supported by your graduate mentor, who specializes in these topic areas and will provide guidance throughout the project. You will also be supported by a group of 8 developers of the hydrostatic machine modeling toolset that are working on different aspects of the code.
Research categories:
Big Data/Machine Learning, Deep Learning, Fluid Modelling and Simulation, Material Modeling and Simulation
Preferred major(s):
- No Major Restriction
Desired experience:
• Intermediate knowledge of fluid mechanics
• Basic programming knowledge (Python, MATLAB, C++, or similar languages)
School/Dept.:
Mechanical Engineering
Professor:
Lizhi
Shang
More information: https://engineering.purdue.edu/Maha/
Mixed-Reality Testbed for Human-Robot Interaction
Description:
In various emerging applications of autonomy, including autonomous driving, teleoperation, and assistive robotics, a human and an autonomous system closely interact with each other. This project’s goal is to develop a mixed-reality platform to facilitate research on trustworthy human-robot interaction. The platform will enable the creation of different environments and scenarios, in which human users/operators and robots interact, to facilitate the development and training of learning-based control algorithms for autonomous systems. The undergraduate researcher will contribute to setting up the mixed-reality platform as well as the design and implementation of high-level control algorithms.
Research categories:
Big Data/Machine Learning, Fabrication and Robotics, Human Factors
Preferred major(s):
- No Major Restriction
School/Dept.:
Electrical and Computer Engineering
Professor:
Mahsa
Ghasemi
More information: https://mahsaghasemi.github.io/
Molecular microscopy to inform the design of medications
Description:
As illustrated with the COVID vaccines, storage and stability of medications can limit widespread availability. We are developing innovative chemical imaging tools with ultrafast pulsed lasers capable of mapping transformations within medical formulations to model and inform stability and bioavailability. Depending on the interests of the students, project scope can range from: i) bench-science in sample preparation and characterization, ii) instrument design and optical path alignment, iii) data acquisition and image analysis algorithm development, iv) partnership with collaborators in the pharmaceutical industry. We have a vibrant and diverse cohort of current researchers dedicated to fostering a supportive and collaborative research environment for all.
Research categories:
Big Data/Machine Learning, Biological Characterization and Imaging, Material Processing and Characterization, Medical Science and Technology
Preferred major(s):
- No Major Restriction
School/Dept.:
Chemistry
Professor:
Garth
Simpson
More information: http://www.chem.purdue.edu/simpson/
Nanoscale High-Speed 3D Printing
Description:
The ability to create 3D structures in the micro and nanoscale is important for many applications including electronics, microfluidics, and tissue engineering. This project deals with development and testing of a setup for building 3D structures using a femtosecond pulsed laser. A method known as two photon polymerization is typically used to fabricate such structures in which a polymer is exposed to a laser beam and at the point of the exposure the polymer changes its structure. Moving the laser in a predefined path helps in getting the desired shape, and the structures are then built in a layer by layer fashion. The setup incorporates all the steps from a designing a CAD model file to slicing the model in layers to generating the motion path of the laser needed for fabricating the structure. Like many other 3D printing processes, 3D printing at nanoscale is also slow. In order to make a 3D structure rapidly, many processes are currently being developed, including projecting 2D images and printing 3D structures in a rapid, layer-by-layer fashion. Other efforts include the use of machine learning to produce high quality 3D parts and printing materials other than polymers to achieve specific mechanical, electrical or optical properties. The undergraduate student will work with graduate student to learn the state-of-the-art 3D nanoprinting systems, help to develop rapid printing processes, and analyze printing results.
Research categories:
Big Data/Machine Learning, Deep Learning, Fabrication and Robotics, Material Processing and Characterization, Nanotechnology
Preferred major(s):
- Mechanical Engineering
- Physics
- Industrial Engineering
- Computer Engineering
Desired experience:
Junior or Senior standing, knowledge in CAD, knowledge in Python is a plus
School/Dept.:
Mechanical Engineering
Professor:
Xianfan
Xu
More information: https://engineering.purdue.edu/~xxu/; https://engineering.purdue.edu/NanoLab/
Non-Invasive Physiological Signals that Indicate Severity of Parkinsons Disease
Description:
The goal is to collect data non-invasively from physiological signals that can be used as indicators of disorders in the motor cortex in subjects. Additionally, we hope to relate any abnormalities in these signals back to past behaviours and experiences such as demographics (e.g. age and gender), medication and head injuries (TBI). The aim for the summer is to create and test methods for obtaining a reliable, non-invasive measure of galvanic skin response from patients as well as to obtain patient test data from Parkinson's Disease at IU Health Physicians Neurology clinic in Indianapolis.
These methods will first replicate and then test conditions important for measuring motor function and size effects. Findings will be used as pilot data for possible future research.
These methods will first replicate and then test conditions important for measuring motor function and size effects. Findings will be used as pilot data for possible future research.
Research categories:
Big Data/Machine Learning, Medical Science and Technology
Preferred major(s):
- Biomedical Engineering
Desired experience:
Human Subject Data Collection Experience
Statistics
Signal Processing
School/Dept.:
Psychological Sciences
Professor:
Anne
Sereno
Physics-Informed Machine Learning to Improve the Predictability of Extreme Weather Events
Description:
Atmospheric blocking events and 'Bomb Cyclones' are an important contributor to high impact extreme weather events. Both these weather extremes lead to heat waves, cold spells, droughts, and heavy precipitation episodes, which have dire consequences for the public health, economy, and ecosystem. For example, the blocking-induced heat waves of 2003 in Europe led to tens of thousands of human casualties and tens of billions of dollars of financial damage.
Traditionally, prediction of extreme weather events is based on direct numerical simulation of regional or global atmospheric models, which are expensive to conduct and involve a large number of tunable parameters. However, with the rapid rise of data science and machine learning in recent years, this proposed work will apply convolutional neural network to an idealized atmospheric model to conduct predictability analysis of extreme weather events within this model. With this proposed machine-learning algorithm, our project will provide a robust forecast of heat waves and atmospheric blocking with a lead-time of a few weeks. With more frequent record-breaking heat waves in the future, such a prediction will offer a crucial period of time (a few weeks) for our society to take proper preparedness steps to protect our vulnerable citizens.
This project is based on developing and verifying the machine learning algorithm for detecting extreme weather events in an idealized model. We will use Purdue’s supercomputer Bell to conduct the simulations. The undergraduate student will play an active and important role in running the idealized model, and participate in developing the algorithms. As an important component of climate preparedness, the proposed work aims to develop a physics-informed machine learning framework to improve predictability of extreme weather events.
Closely advised by Prof. Wang, the student will conduct numerical simulations of an idealized and very simple climate model, and use python-based machine learning tools to predict extreme weather events within the model. Prof. Wang will provide weekly tutorial sessions to teach key techniques along with interactive hands-on sessions. The students will get access to the big datasets on Purdue’s Data Depot, analyze and visualize data of an idealized atmospheric model. The student will use convolutional neural networks (CNNs) to train and assess a Machine-Learning model. The student will further use feature tracking algorithm to backward identify the physical structure in the atmosphere that is responsible for the onset of extreme weather events.
Traditionally, prediction of extreme weather events is based on direct numerical simulation of regional or global atmospheric models, which are expensive to conduct and involve a large number of tunable parameters. However, with the rapid rise of data science and machine learning in recent years, this proposed work will apply convolutional neural network to an idealized atmospheric model to conduct predictability analysis of extreme weather events within this model. With this proposed machine-learning algorithm, our project will provide a robust forecast of heat waves and atmospheric blocking with a lead-time of a few weeks. With more frequent record-breaking heat waves in the future, such a prediction will offer a crucial period of time (a few weeks) for our society to take proper preparedness steps to protect our vulnerable citizens.
This project is based on developing and verifying the machine learning algorithm for detecting extreme weather events in an idealized model. We will use Purdue’s supercomputer Bell to conduct the simulations. The undergraduate student will play an active and important role in running the idealized model, and participate in developing the algorithms. As an important component of climate preparedness, the proposed work aims to develop a physics-informed machine learning framework to improve predictability of extreme weather events.
Closely advised by Prof. Wang, the student will conduct numerical simulations of an idealized and very simple climate model, and use python-based machine learning tools to predict extreme weather events within the model. Prof. Wang will provide weekly tutorial sessions to teach key techniques along with interactive hands-on sessions. The students will get access to the big datasets on Purdue’s Data Depot, analyze and visualize data of an idealized atmospheric model. The student will use convolutional neural networks (CNNs) to train and assess a Machine-Learning model. The student will further use feature tracking algorithm to backward identify the physical structure in the atmosphere that is responsible for the onset of extreme weather events.
Research categories:
Big Data/Machine Learning, Deep Learning, Energy and Environment, Environmental Characterization, Fluid Modelling and Simulation
Preferred major(s):
- Physics
- Planetary Sciences
- Atmospheric Science/Meteorology
- Computer Science
- Mathematics - Computer Science
- Mathematics
- Environmental Geosciences
- Mechanical Engineering
- Civil Engineering
- Aeronautical and Astronautical Engineering
- Computer Engineering
- Engineering (First Year)
- Multidisciplinary Engineering
- Natural Resources and Environmental Science (multiple concentrations)
Desired experience:
Familiar with Machine Learning or prior knowledge of convolutional neural networks (CNNs);
Have basic level training on PHYS172 Modern Mechanics or PHYS 15200 Mechanics or equivalent courses from other institutions; Familiar with Python scripting and visualization
School/Dept.:
Earth, Atmospheric, and Planetary Sciences
Professor:
Lei
Wang
More information: https://www.eaps.purdue.edu/people/profile/wanglei.html
Polaritonic Energy Transport: Hybridizing Radiation and Conduction for Microelectronics Cooling
Description:
Who we are… Specere is a latin word that means “to look or behold.” That’s what we do. We look, explore, and examine different ways to: (1) move energy with light and (2) get information from light. More specifically, we are a light lab employing infrared physics to create spectroscopic, thermal, and sensing solutions.
Who we are seeking… We look for motivated and hard-working undergraduates having both strong aspirations for post-graduate studies as well as those that are just “grad school curious.” All applicants should be capable of working independently while effectively communicating within a team setting.
Research Topic, Polaritonic Energy Transport: We seek to design materials capable of more effectively moving heat at extremely small scales like those in modern microelectronics. Success will enable: more efficient data centers, power electronics like those in EV’s, and new computing architectures.
What’ You’ll Do: Team members will be responsible for designing novel metamaterial stacks capable of maximizing heat transfer using a combination of computational modeling and experimental measurements of optical properties. Direct mentoring from Dr. Beechem will build your skills up in each area such that you will gain proficiency in advanced simulation (COMSOL) and spectroscopic tools (Raman, IR-ellipsometry). In addition, you will have the chance to participate in writing journal articles and pursuing patents based on your work.
Who we are seeking… We look for motivated and hard-working undergraduates having both strong aspirations for post-graduate studies as well as those that are just “grad school curious.” All applicants should be capable of working independently while effectively communicating within a team setting.
Research Topic, Polaritonic Energy Transport: We seek to design materials capable of more effectively moving heat at extremely small scales like those in modern microelectronics. Success will enable: more efficient data centers, power electronics like those in EV’s, and new computing architectures.
What’ You’ll Do: Team members will be responsible for designing novel metamaterial stacks capable of maximizing heat transfer using a combination of computational modeling and experimental measurements of optical properties. Direct mentoring from Dr. Beechem will build your skills up in each area such that you will gain proficiency in advanced simulation (COMSOL) and spectroscopic tools (Raman, IR-ellipsometry). In addition, you will have the chance to participate in writing journal articles and pursuing patents based on your work.
Research categories:
Big Data/Machine Learning, Material Modeling and Simulation, Material Processing and Characterization, Nanotechnology, Thermal Technology
Preferred major(s):
- No Major Restriction
Desired experience:
Proficiency in Matlab, COMSOL or both is a plus.
School/Dept.:
School of Mechanical Engineering
Professor:
Thomas
Beechem
More information: www.specere.org
Real time analysis of viral particles
Description:
The increasing worldwide demand for vaccines along with the intensifying economic pressure on health care systems underlines the need for further improvement of vaccine manufacturing. In addition, regulatory authorities are encouraging investment in continuous manufacturing process to ensure robust production, avoid shortages, and ultimately lower the cost of medications for patients. The limitations of in-line process analytical tools are a serious drawback of the efforts taken in place. In line analysis of viral particles are very limited, due to the large time required for the current techniques for detection, qualitative and quantitative analysis. Therefore, there is a need for new process analytical technology. This project has both experimental and computation components and two students will be recruited for perform different tasks. The student focusing on experiments will fabricate devices and test them. The student focusing on computations will focus on developing machine learning codes.
Research categories:
Big Data/Machine Learning, Biological Characterization and Imaging, Biological Simulation and Technology, Biotechnology Data Insights, Cellular Biology, Computer Architecture, Deep Learning
Preferred major(s):
- No Major Restriction
Desired experience:
For the experimental portion of the project: fabrication, cell culture, microfluidics, microscopy
For the computational portion of the project: Coding, Python
School/Dept.:
Mechanical Engineering
Professor:
Arezoo
Ardekani
More information: https://engineering.purdue.edu/ComplexFlowLab/
Real-Time Measurements of Volatile Chemicals in Buildings with Proton Transfer Reaction Mass Spectrometry
Description:
The objective of this project is to utilize state-of-the-art proton transfer reaction mass spectrometry (PTR-MS) to evaluate emissions and exposures of volatile chemicals in buildings. My group is investigating volatile chemical emissions from consumer and personal care products, disinfectants and cleaning agents, and building and construction materials. You will assist graduate students with full-scale experiments with our PTR-MS in our new Purdue zEDGE Tiny House and process and analyze indoor air data in MATLAB.
Research categories:
Big Data/Machine Learning, Ecology and Sustainability, Energy and Environment, Engineering the Built Environment, Environmental Characterization, Human Factors, Internet of Things
Preferred major(s):
- No Major Restriction
Desired experience:
Preferred skills: experience with MATLAB, Python, or R. Coursework: environmental science and chemistry, physics, thermodynamics, heat/mass transfer, and fluid mechanics.
School/Dept.:
Lyles School of Civil Engineering
Professor:
Nusrat
Jung
More information: https://www.purdue.edu/newsroom/stories/2020/Stories%20at%20Purdue/new-purdue-lab-provides-tiny-home-for-sustainability-education.html
Renewable energy-powered water technologies
Description:
Water and energy are tightly linked resources that must both become renewable for a successful future. However, today, water and energy resources are often in conflict with one another, especially related to impacts on electric grids. Further, advances in nanotechnology, material science and artificial intelligence allow for new avenues to improve the widespread implementation of desalination and water purification technology. The team is pursuing multiple projects that aim to explore solar and wind-powered desalination, nanofabricated membranes, light-driven reactions, artificial intelligence control algorithms, and thermodynamic optimization of energy systems. The student will be responsible for fabricating membranes, building hydraulic systems, modeling thermal fluid phenomenon, analyzing data, or implementing control strategies in novel system configurations. More information here: www.warsinger.com
Research categories:
Big Data/Machine Learning, Chemical Catalysis and Synthesis, Ecology and Sustainability, Energy and Environment, Engineering the Built Environment, Environmental Characterization, Fluid Modelling and Simulation, Material Modeling and Simulation, Nanotechnology, Thermal Technology
Preferred major(s):
- Mechanical Engineering
- Civil Engineering
- Environmental and Ecological Engineering
- Chemistry
- Chemical Engineering
- Materials Engineering
Desired experience:
Applicants should have an interest in thermodynamics, water treatment, and sustainability. Applicants with experience in some (not all) of the following are preferred: experimental design and prototyping, manufacturing, Python, LabView, EES, MATLAB, 3D CAD Software, & Adobe Illustrator. Rising Juniors and Seniors are preferred.
School/Dept.:
Mechanical Engineering
Professor:
David
Warsinger
More information: www.warsinger.com
Resilient AI Network (RAIN) for next-generation manufacturing.
Description:
Private project for Devin Kelly
Research categories:
Big Data/Machine Learning
Preferred major(s):
- No Major Restriction
School/Dept.:
ECE
Professor:
Ali
Shakouri
Robust Deep Learning in Complex Real-World Environment
Description:
Machine learning discovers statistical knowledge from data and has escaped from the cage of perception. A growing number of complex systems from walking robots, drones to the computer Go player rely on learning techniques to make decisions to achieve optimal control of complex systems. This change represents a truly fundamental departure from traditional classification and regression methods as such learning systems must deal with complex real-world environment. This team investigates how to utilize robust deep learning for predicting complex real-world environment, such as autonomous driving, and complex network systems.
Research categories:
Big Data/Machine Learning
Preferred major(s):
- No Major Restriction
Desired experience:
Familiar with Python programming and PyTorch or TensorFlow. Have some basic knowledge about machine learning and neural network.
School/Dept.:
School of Mechanical Engineering
Professor:
Guang
Lin
Sentiment analysis and performance monitoring of the air transport sector
Description:
The aim of this project is to build a pipeline to automate the data collection, analysis, and visualization process so that we can monitor the customer satisfaction of the air transport sector. The student researcher will be responsible for the design and implementation of the project with the supervision and guidance of the faculty advisor.
Research categories:
Big Data/Machine Learning
Preferred major(s):
- No Major Restriction
Desired experience:
The student should have good motivation for learning.
The student should have some skills and experience of coding.
The student need to have some general understanding of statistics.
The student is expected to communicate with the faculty advisor often during the project period.
School/Dept.:
School of Aviation and Transportation Technology
Professor:
Yi
Gao
Software for deep learning and deep learning for software
Description:
Possible industry involvement: Some of these projects are funded by Facebook research awards and J.P.Morgan AI research awards.
We have three openings for different tasks including those listed below.
NOTE: We especially encourage applications from women, Aboriginal peoples, and other groups underrepresented in computing.
*** Subproject 1. Testing Deep Learning Systems
We will build cool and novel techniques to make deep learning code such as TensorFlow and PyTorch reliable and secure. We will build it on top of our award-winning paper (ACM SIGSOFT Distinguished Paper Award)!
Machine learning systems including deep learning (DL) systems demand reliability and security. DL systems consist of two key components: (1) models and algorithms that perform complex mathematical calculations, and (2) software that implements the algorithms and models. Here software includes DL infrastructure code (e.g., code that performs core neural network computations) and the application code (e.g., code that loads model weights). Thus, for the entire DL system to be reliable and secure, both the software implementation and models/algorithms must be reliable and secure. If software fails to faithfully implement a model (e.g., due to a bug in the software), the output from the software can be wrong even if the model is correct, and vice versa.
This project aims to use novel approaches including differential testing to detect and localize bugs in DL software (including code and data) to address the testing oracle challenge.
Early work and background can be found here:
EAGLE: Creating Equivalent Graphs to Test Deep Learning Libraries (our ICSE 2022 paper, forthcoming, check my homepage)
https://www.cs.purdue.edu/homes/lintan/publications/fairness-neurips21.pdf
https://www.cs.purdue.edu/homes/lintan/publications/variance-ase20.pdf
https://www.cs.purdue.edu/homes/lintan/publications/cradle-icse19.pdf
*** Subproject 2. Inferring Specifications from Software Text for Finding Bugs and Vulnerabilities
A fundamental challenge of detecting or preventing software bugs and vulnerabilities is to know programmers’ intentions, formally called specifications. If we know the specification of a program (e.g., where a lock is needed, what input a deep learning model expects, etc.), a bug detection tool can check if the code matches the specification.
Building upon our expertise on being the first to extract specifications from code comments to automatically detect software bugs and bad comments, in this project, we will analyze various new sources of software textual information (such as API documents and StackOverflow Posts) to extract specifications for bug detection. For example, the API documents of deep learning libraries such as TensorFlow and PyTorch contain a lot of input constraint information about tensors.
Early work and background can be found here:
https://www.cs.purdue.edu/homes/lintan/projects.html
*** Subproject 3. Leveraging Deep Learning to Detect and Fix Software Bugs and Vulnerabilities
In this project, we will develop cool machine learning approaches to automatically learn bug and vulnerability patterns and fix patterns from historical data to detect and fix software bugs and security vulnerabilities.
Early work and background can be found here:
https://www.cs.purdue.edu/homes/lintan/publications/cure-icse21.pdf
https://www.cs.purdue.edu/homes/lintan/publications/deeplearn-tse18.pdf
We have three openings for different tasks including those listed below.
NOTE: We especially encourage applications from women, Aboriginal peoples, and other groups underrepresented in computing.
*** Subproject 1. Testing Deep Learning Systems
We will build cool and novel techniques to make deep learning code such as TensorFlow and PyTorch reliable and secure. We will build it on top of our award-winning paper (ACM SIGSOFT Distinguished Paper Award)!
Machine learning systems including deep learning (DL) systems demand reliability and security. DL systems consist of two key components: (1) models and algorithms that perform complex mathematical calculations, and (2) software that implements the algorithms and models. Here software includes DL infrastructure code (e.g., code that performs core neural network computations) and the application code (e.g., code that loads model weights). Thus, for the entire DL system to be reliable and secure, both the software implementation and models/algorithms must be reliable and secure. If software fails to faithfully implement a model (e.g., due to a bug in the software), the output from the software can be wrong even if the model is correct, and vice versa.
This project aims to use novel approaches including differential testing to detect and localize bugs in DL software (including code and data) to address the testing oracle challenge.
Early work and background can be found here:
EAGLE: Creating Equivalent Graphs to Test Deep Learning Libraries (our ICSE 2022 paper, forthcoming, check my homepage)
https://www.cs.purdue.edu/homes/lintan/publications/fairness-neurips21.pdf
https://www.cs.purdue.edu/homes/lintan/publications/variance-ase20.pdf
https://www.cs.purdue.edu/homes/lintan/publications/cradle-icse19.pdf
*** Subproject 2. Inferring Specifications from Software Text for Finding Bugs and Vulnerabilities
A fundamental challenge of detecting or preventing software bugs and vulnerabilities is to know programmers’ intentions, formally called specifications. If we know the specification of a program (e.g., where a lock is needed, what input a deep learning model expects, etc.), a bug detection tool can check if the code matches the specification.
Building upon our expertise on being the first to extract specifications from code comments to automatically detect software bugs and bad comments, in this project, we will analyze various new sources of software textual information (such as API documents and StackOverflow Posts) to extract specifications for bug detection. For example, the API documents of deep learning libraries such as TensorFlow and PyTorch contain a lot of input constraint information about tensors.
Early work and background can be found here:
https://www.cs.purdue.edu/homes/lintan/projects.html
*** Subproject 3. Leveraging Deep Learning to Detect and Fix Software Bugs and Vulnerabilities
In this project, we will develop cool machine learning approaches to automatically learn bug and vulnerability patterns and fix patterns from historical data to detect and fix software bugs and security vulnerabilities.
Early work and background can be found here:
https://www.cs.purdue.edu/homes/lintan/publications/cure-icse21.pdf
https://www.cs.purdue.edu/homes/lintan/publications/deeplearn-tse18.pdf
Research categories:
Big Data/Machine Learning, Cybersecurity, Deep Learning, Other
Preferred major(s):
- Computer Science
- Computer Engineering
- Software Engineering
Desired experience:
Strong coding skills and motivation in research are required. Background in security or machine learning is not required but a plus.
School/Dept.:
Computer Science
Professor:
Lin
Tan
More information: https://www.cs.purdue.edu/homes/lintan/research.html
Spatio-temporal modeling of inter-correlated arrivals: A case study on Indiana state-wide nursing homes.
Description:
Maintaining an appropriate staffing level is essential to providing a healthy workplace
environment at nursing homes and ensuring quality care among residents. To alleviate staffing level concerns and maintain the necessary care quantity, we use the Minimum Data Set (MDS) to calculate and forecast the resident arrivals and care needs (measured in staff-time). For the care transition from hospitals to nursing homes, the demand come from different hospitals and dispose to different nursing facilities. Thus, there is a need to analyze the spatial and temporal pattern of the inter-correlated residents arrivals, confounded by multi-level risk factors, including individual level (clinical and behavioral assessment variables), health systems level (access to acute care), and health policy level.
environment at nursing homes and ensuring quality care among residents. To alleviate staffing level concerns and maintain the necessary care quantity, we use the Minimum Data Set (MDS) to calculate and forecast the resident arrivals and care needs (measured in staff-time). For the care transition from hospitals to nursing homes, the demand come from different hospitals and dispose to different nursing facilities. Thus, there is a need to analyze the spatial and temporal pattern of the inter-correlated residents arrivals, confounded by multi-level risk factors, including individual level (clinical and behavioral assessment variables), health systems level (access to acute care), and health policy level.
Research categories:
Big Data/Machine Learning
Preferred major(s):
- Economics (School of Management)
- Computer Science
- Data Science
- Industrial Engineering
Desired experience:
basic data processing courses offered by the Data Mine or equivalent;
some experience in processing and visualizing large amount of data, preferably healthcare claims data.
basic knowledge in data modeling (e.g., regression), preferably time-series analysis and spatial analysis
School/Dept.:
Biomedical Engineering
Professor:
Nan
Kong
Studying the role of infection in specific cancers
Description:
We are interested in identifying how certain viruses are causing specific cancers. To this end, we hope to extract host and pathogen specific gene expression patterns from single cell omics technologies from public data.
Research categories:
Big Data/Machine Learning, Cellular Biology
Preferred major(s):
- No Major Restriction
Desired experience:
familiarity with genomics technologies are welcome
School/Dept.:
Computer Science
Professor:
Majid
Kazemian
More information: kazemianlab.com
Surface sound and AI based machine monitoring for smart manufacturing
Description:
A stethoscope-based surface sound sensor has been developed at Purdue and this sensor is being used to monitor manufacturing equipment. Deep learning will be used to classify machine and process conditions using the sound sensor. Student will work with PhD student to implement sound sensors on machines in local manufacturing companies, collect data, build database and dashboard, and develop AI models.
Research categories:
Big Data/Machine Learning, Deep Learning, Fabrication and Robotics, Internet of Things
Preferred major(s):
- Mechanical Engineering
- Computer Engineering
- Electrical Engineering
- Computer and Information Technology
- Computer Science
School/Dept.:
Mechanical Engineering
Professor:
Martin
Jun
More information: https://web.ics.purdue.edu/~jun25/
Thermal management of electronic devices
Description:
The continued miniaturization of electronic devices, with expanded functionality at reduced cost, challenges the viability of products across a broad spectrum of industry applications. The electronics industry is driven by global trends in storage, transmission, and processing of extreme quantities of digital information (cloud computing, data centers), increasing electrification of the transportation sector (electric vehicles, hybrid aircraft, batteries), and the proliferation of interconnected computing devices (mobile computing, IoT, 5G). Proper thermal management of electronic devices is critical to avoid overheating failures and ensure energy efficient operation. In view of these rapidly evolving markets, most of the known electronics cooling technologies are approaching their limits and have a direct impact on system performance (e.g., computing power, driving range, device size, etc.).
Research projects in the Cooling Technologies Research Center (CTRC) are exploring new technologies and discovering ways to more effectively apply existing technologies to addresses the needs of companies and organizations in the area of high-performance heat removal from compact spaces. One of the distinctive features of working in this Center is training in practical applications relevant to industry. All of the projects involve close industrial support and collaboration in the research, often with direct transfer of the technologies to the participating industry members. Projects in the Center involve both experimental and computational aspects, are multi-disciplinary in nature, and are open to excellent students with various engineering and science backgrounds. Multiple different research project opportunities are available based on student interests and preferences.
Research projects in the Cooling Technologies Research Center (CTRC) are exploring new technologies and discovering ways to more effectively apply existing technologies to addresses the needs of companies and organizations in the area of high-performance heat removal from compact spaces. One of the distinctive features of working in this Center is training in practical applications relevant to industry. All of the projects involve close industrial support and collaboration in the research, often with direct transfer of the technologies to the participating industry members. Projects in the Center involve both experimental and computational aspects, are multi-disciplinary in nature, and are open to excellent students with various engineering and science backgrounds. Multiple different research project opportunities are available based on student interests and preferences.
Research categories:
Big Data/Machine Learning, Energy and Environment, Fluid Modelling and Simulation, Material Modeling and Simulation, Nanotechnology, Thermal Technology
Preferred major(s):
- No Major Restriction
School/Dept.:
School of Mechanical Engineering
Professor:
Justin
Weibel
More information: https://engineering.purdue.edu/CTRC/research/
Transport in Vanadium Oxides
Description:
Vanadium oxides undergo a metal-insulator transition, changing their resistivity by five orders of magnitude. However, they don't do it all at once! Rather, the material changes piece by piece, at the nanoscale. This project will be to model in detail the total resistance of a substance that has interleaved bits of metal and insulator.
Research categories:
Big Data/Machine Learning, Material Modeling and Simulation
Preferred major(s):
- Physics
Desired experience:
Experience programming and in data analysis is strongly preferred.
School/Dept.:
Physics and Astronomy
Professor:
Erica
Carlson
More information: http://www.physics.purdue.edu/~erica/
Understanding Quantum Correlations of Light for Imaging
Description:
We aim to leverage the expertise in two fields of computational and quantum imaging to develop classical algorithms to optimize and process quantum correlated images. On one hand, we introduce quantum complexity to imaging algorithms which deserves the attention of AI-assisted signal/image processing to extract hidden information from measurements. On the other hand, we iteratively engineer quantum states of a light source to enhance imaging resolution. Our goal is to implement a room-temperature quantum light source and understand and optimize its quantum correlations in multiple dimensions. We plan to apply computational and machine learning methods to reconstruct images using model-based gradient ascent and Bayesian estimation techniques.
Research categories:
Big Data/Machine Learning, Material Processing and Characterization, Nanotechnology, Thermal Technology
Preferred major(s):
- Electrical Engineering
- Computer Science
- Physics
Desired experience:
Junior or Senior students with experience/knowledge of image processing, machine learning and optics. GPA>3.5
School/Dept.:
Electrical and computer Engineering
Professor:
Mahdi
Hosseini
Vision Based Navigation for UAVs
Description:
Develop optical flow/depth estimation with dynamic vision sensor cameras as the only vision sensor.
Research categories:
Big Data/Machine Learning, Deep Learning
Preferred major(s):
- Electrical Engineering
Desired experience:
Deep learning
School/Dept.:
IN
Professor:
Kaushik
Roy