Research Statement

Current Projects (PNW):

Early Concept Exploratory Summer Research: Improvement of Arcelor Mittal Mill Operation through Incorporation of Data Fusion and Machine Learning (2019-present)

  • Overall premise: to apply machine learning algorithms and data fusion techniques to a variety of ArcelorMittal mill operations
  • Serves as graduate thesis topic for one MS student as well as ongoing research for PIs

Data Visualization Tools for High School Machine Learning Education (2019-present)

  • Overall premise: to develop data visualization tools for machine learning education with emphasis on high school students
  • Serves as graduate thesis topic for one MS student

Adaptable Intramuscular Injection Device (2019-2020)

  • Overall premise: to create the first-generation prototype of an adaptable intramuscular injection device that would ease this process for patients who self-inject medications intramuscularly at home
  • Serves as senior design project for three undergraduate mechanical engineering students

Past Projects (PNW):

Augmented Reality Scavenger Hunt (2018-2019)

  • Overall premise: to design an augmented reality Android application that guides prospective students around ECE labs and facilities using ECE-based trivia questions
  • Served as hourly research project to four undergraduate students

Prediction of Cognitive Workload Using Machine Learning Analysis of Physiological Data (2018-2019)

  • Overall premise: to predict in real time early warnings of cognitive overload or underload in military (e.g. pilots, cyber security threat detection) and civilian (e.g. driver attentiveness) applications
  • Served as senior design project for two undergraduate students

Data Acquisition to Determine Cognitive Workload (2018-2019)

  • Overall premise: to design a smartwatch capable of capturing physiological data to use in conjunction with the above project
  • Served as senior design project for two undergraduate students

Past Projects (UT):

Assessment of Team Dynamics Using Adaptive Modeling of Biometric Data (2016-2018)

  • Overall premise: to analyze, compare, and enhance a variety of machine learning algorithms for assessment of cognitive workload in human factors based applications using MATLAB and Python
  • My end of the project: due to the nature of the project being a fellowship based on a student-faculty partnership, my end consists of all technical and analytical aspects from start to finish

Sliding Scale Autonomy through Physiological Rhythm Evaluations (SAPHYRE) (2016-2018)

  • Overall premise: to form classifications and predictions of pilot skill and workload using flight-based and physiological data and machine learning algorithms in Python
  • My end of the project: to supervise and provide guidance to an M.S. student whose thesis consisted heavily of this work

Episodic Memory Reconstruction for UAV Behavior Explanation (EpEx) (2016-2018)

  • Overall premise: to analyze and explain the behavior of autonomous UAVs through past and present sensory data stored in episodic memory
  • End objective: to utilize this improved explainable behavior to minimize discrepancies between predictions and outcomes of UAV behavior, thereby increasing chances of mission success in military and civilian applications
  • My end of the project: to design and implement machine learning algorithms for prediction of UAV data through a combination of sensor fusion and artificial neural networks using MATLAB

I-Corps Teams: Wireless Sensor Network Based Localization and Navigation for Precision Agriculture (2016-2017)

  • Overall premise: to develop a hardware implementation of the wireless sensor network algorithm developed in the EAGER project (described below) in the application of precision agriculture
  • Over time, based on extensive feedback from potential customers, the focus shifted to an application of logistics and warehouse navigation with broader technological frameworks
  • Conducted extensive research studies on large volumes of potential customers throughout nearby industrial cities, including Toledo, Chicago, Pittsburgh, Philadelphia, and Long Island
  • My end of the project: to supply the initial localization algorithm and to assist another PhD student (the entrepreneurial lead) with writing and customer survey related work

EAGER: Localization in Ad-Hoc Wireless Networks: Investigation into Fusing Dempster-Shafer Theory and Support Vector Machines (2014-2016)

  • Overall premise: to develop low cost localization algorithms for wireless sensor networks by fusing Dempster-Shafer theory and support vector machines
  • End objective: to enhance the accuracy and computational efficiency of sensor network localization in time-critical applications
  • My end of the project: to incorporate Dempster-Shafer theory into a localization algorithm developed by an outgoing M.S. student, which resulted in up to 97% accuracy in a fraction of the runtime required by previous established localization techniques