Updated on April 1, 2020
Several open positions are available for PhD students at Velni’s Lab in the School of ECE at Univ. of Georgia. The positions are available as early as August 2020. The topics and the desired background are as follows:
1. Learning-based and Uncertainty-aware Control for Complex Systems
- The objective of this project is to develop fundamental tools for real-time (and data-driven) model learning and predictive control of nonlinear and stochastic systems, where uncertainties are also learned, and apply them to several practical applications. The modeling and control design would be done in the linear parameter-varying (LPV) framework.
- Strong background in machine learning and model predictive control (MPC) is required. Prior knowledge of LPV systems modeling and/or control is desired.
2. Development of a Big Data Analytics Pipeline for Precision Agriculture
- The general goal of this project is to utilize and implement real-time learning tools for a variety of applications in smart farming, and in particular in high-throughput phenotyping.
- Strong background in machine/deep learning theory and application and big data analytics is required.
3. Stochastic Hybrid Control Design for Mass Deployment of Autonomous Vehicles
- The general goal of the project is to develop a model-based stochastic hybrid systems theoretic approach to coordinate a large group of connected and automated vehicles. In particular, the project will develop a stochastic MPC framework accounting for probabilistic uncertainties in disturbances (e.g., human actions) and vehicles’ operating mode transitions.
- Strong background in hybrid control systems is required. Prior knowledge of statistical machine learning is desired.
4. Anomaly Detection in Cyber-human-physical Systems using Statistical Learning
- The general goal of the project is to develop distributed approaches for detecting anomalies (faults or attacks) in cyber-physical systems, and in particular in smart energy systems. The development of theory and implementation of distributed real-time learning approaches is the goal, where the prior knowledge of the underlying physics-based models will be embedded in the appropriate data-driven method and used for detecting anomalies.
- Strong background in graph theory and machine learning is required. Prior knowledge of domain (e.g., power systems) is desired.
A competitive research assistantship and full tuition waiver will be offered. Outstanding candidates will also be considered for other fellowships provided by the UGA College of Engineering and Graduate School.
To apply, please send an application package to email@example.com. The application should be submitted as a single PDF and include a cover letter (explicitly describing the candidate background and how they fit the open positions), a detailed CV (including the list of publications), and unofficial copies of their BS (and, if applicable, MS) transcripts. Due to the volume of emails I receive in response to these open positions, I may not be able to respond to all enquiries, for which I apologize.
University of Georgia (UGA), a top tier one research institution, is ranked 16th overall among all public national universities in the 2018 U.S. News & World Report rankings, and a Princeton Review top ten in value. UGA is recognized as a Public Ivy, a publicly-funded university considered to provide a quality of education comparable to that of an Ivy League university. Athens, GA is located approximately 70 miles northeast of Atlanta, GA. Consistently voted one of the best college towns in the United States, Athens has a thriving business, restaurant and music scene and is the gateway to numerous leisure activities in northern Georgia.