First Year Projects
Johnathan Brantley – Department of Chemistry
Cyclic Cumulenes as Enabling Motifs in Functional Materials
Polymers that contain cyclic repeating units are important synthetic targets, given they often exhibit unique physical properties. Brantley’s project will explore vinyl-addition polymerizations of cyclic allenes to access materials with new properties and potential applications.
Subhadeep Chakraborty – Department of Mechanical, Aerospace and Biomedical Engineering
Artificial Intelligence based impairment detection system for vehicle operators through combined analysis of physiological and traffic sensor data
Impaired driving is a key contributing factor leading to more than 10,000 fatalities in 2016. By integrating and fusing multiple data sources such as driver biometrics, vehicle kinematics, and roadway and environmental conditions in real-time, this project aims to generate an intelligent Advanced Driver Assist System (iADAS) which will provide useful feedback to drivers and potentially mitigate accidents.
Jian Liu – Min H. Kao Department of Electrical Engineering and Computer Science
Towards Robust and Trustworthy Federated Learning for Ubiquitous Cyber-Physical Systems: Security, Privacy, and Scalability
Different from traditional centralized training, federated learning distributes the training process to the edge, enabling edge-computing devices to collaboratively learn/update a shared model using the data that is kept locally on the device. Liu’s project hopes to build a foundation for understanding how to push AI gains in performance, robustness, and scalability to CPS in mobile edge computing.
Second Year Projects
Mahshid Ahmadi – Department of Materials Science and Engineering
Machine Learning Driven Experimental Approach for the Discovery of New Organic-Inorganic Halide Perovskites for Optoelectronic Applications
Organic-inorganic halide perovskites have emerged as materials of choice for low-cost photovoltaics and optoelectronics due to relatively easy solution synthesis and unique spectrum of functional properties. The objective of this project is to establish a machine learning based experimental approach towards discovery and prediction of hybrid perovskites properties via combinatorial synthesis and evolutionary experiment optimization.
Sindhu Jagadamma – Department of Biosystems Engineering and Soil Science
Is Soil Manganese a Major Driver for Organic Carbon Cycling in Croplands?
Manganese (Mn), an essential plant micronutrient, is believed to play a critical yet poorly understood role in terrestrial ecosystem carbon (C) cycling, particularly in human manipulated agoecosystems. Her team proposes a series of field and laboratory experiments that simulate excess Mn mobilization in agricultural soils to quantitatively assess the role of Mn in ecosystem C cycling. The findings of this proposed study will inform whether acidity-induced elevated Mn availability is a critical driver in soil organic C storage/loss in highly managed agroecosystems.
Hugh Medal – Department of Industrial and Systems Engineering
Machine-Learning-Enabled Modeling for High-Dimensional Dynamics of Materials Processing
Understanding the movement of ions through material is crucial to understanding important properties such as radiation, stress cracking, and ion conductivity. Because of this, there is an urgent need to map high-dimensional energy landscapes (HDEL). Medal’s team plans to address this need via machine-learning-enabled modeling.
Zhenbo Wang – Department of Mechanical, Aerospace and Biomedical Engineering
Real-Time Control for Connected and Automated Vehicles using Traffic Signal
Information
Future roadways will rely on connected and automated vehicles (CAVs) to reduce traffic congestion, maximize fuel economy, and increase safety. Wang’s team plans to develop a novel method to produce the best speed profile using traffic signal phase and timing data. Successful completion of this project would result in a new method for real-time optimal control of CAVs. This work is essential to mainstream the use of CAVs for future transportation systems.
Xiaopeng Zhao – Department of Mechanical, Aerospace and Biomedical Engineering
A Multisensory Brain-Computer Interface for Intelligent Driving
Smart cars and intelligent driving have moved into the forefront of vehicle technology, opening up a number of areas of research. Zhao’s team proposes to investigate intelligent driving through the development of an interface designed to communicate information about a vehicle’s driving conditions to the driver.