loading slideshow...

  • SE-grad-students
  • Pourazarm-2
  • Belta-Schwager-lab
  • SE-VC-posters


Doctoral Students
Setareh Ariafar Advisor: Ioannis Paschalidis
Yuting Chen
I work with Professor Venkatesh Saligrama in the Information Systems and Sciences Lab. My research includes computer vision and statistical learning.  More specifically, we focus on video surveillance problems such as anomalous activity detection, video event retrieval and activity modeling.  We are currently developing algorithms for tasks such as counter-flow detection in airport surveillance videos, multiple people tagging and tracking and event search in airborne surveillance videos. We are trying to solve large-scale problems with simple features, efficient data indexing structure and robust post-processing methods in real time.
Advisor: Venkatesh Saligrama
Evgeniy Goldis
I work with Professor Caramanis to develop a revised day-ahead energy market design that includes flexible, demand-side resources as well as variable generation.  Under the current market structure, both variable generation and demand-side resources do not have incentives to bid their true costs and this biases the market clearing away from the social optimum.  Our proposed design creates appropriate incentives for both participants to bid their respective costs, leading to a socially optimal equilibrium.  I am also working with Professor Caramanis and external team members on a project investigating the benefits of transmission switching in wholesale energy markets.  Initial investigations show that disconnecting certain transmission lines can lead to large savings in production costs.  To that end, we are developing algorithms to identify a set of candidate lines for disconnecting and to quickly select the “best” subset of lines to switch off, while maintaining feasibility of the transmission system.
Advisor: Michael Caramanis
Austin Jones
I work in the Hybrid Networks Systems (HyNeSs) lab with Calin Belta and Mac Schwager on formal synthesis algorithms that involve information that is a priori unknown.  I am integrating techniques in information theory and machine learning with formal methodology to enrich control policies synthesized from specification languages to be able to handle tasks that involve “uncertain” specifications such as exploration or communication.  Ultimately, I plan to use this approach to establish a framework for autonomous, decentralized coordination of teams of mobile robots.
Advisor: Calin Belta
Iman HaghighiI work in the Hybrid and Networked Systems (HyNeSs) lab with Calin Belta. My research interests lie in the fields of control theory and formal methods. Specifically, my current research focuses on using tools and ideas in machine learning, information theory, and control theory in order to learn formal formulae that are able to identify, quantify and ultimately synthesize patterns in distributed and networked systems. We study and develop formal spatial logics that are able to describe static patterns and spatiotemporal logics that deal with dynamic patterns. These logics are further utilized to synthesize system parameters that guarantee emergence of desired patterns. Applications include and are not limited to emergent behaviors in biological networks, patterns produced by swarms of robots, and smart city power management. Advisor: Calin Belta
Emir Kavurmacioglu
My research interests are Network Economics and Wireless Spectrum Management, particularly areas of Dynamic Spectrum Access and Secondary Spectrum Markets, both of which introduce a new way of regulating the telecommunications radio spectrum more efficiently. My current work focuses on analyzing pricing strategies and predicting possible market equilibria when multiple Wireless Spectrum Providers compete and consequently force each other to suboptimal outcomes. This involves the use of statistical tools such as Continuous Time Markov Processes, Dynamic Programming and Linear Optimization as well as the economics field of Game Theory.
Advisor: David Starobinski
Yasaman Khazaeni
I work in the CODES lab with Professor Cassandras. I work in the area of cooperative control. I use the receding horizon method for solving the data harvesting and reward collection problem in sensor networks. My work focuses on optimizing a multiple agent mission in collecting data from many static data generator points. In the optimization problem I try to minimize the total traveling time by the agents while collecting and delivering the data. I use Matlab simulation and will implement the final algorithm on actual robots in our lab.
Advisor: Christos Cassandras
Julia Lima Fleck
I work in the CODES Lab where our research interests encompass the design and control of discrete event and hybrid systems. Given that estimating the performance of Stochastic Hybrid Systems (SHS) is a difficult task, we seek to approximate the relationship between performance metrics and control/design parameters of such systems through the use of Infinitesimal Perturbation Analysis (IPA). We are currently evaluating the many issues involved in applying IPA to Stochastic Flow Models (SFM), which constitute a class of SHS where the time-driven dynamics are captured by flow dynamics and the event-driven dynamics are represented by switches that alter the system’s flow dynamics. In particular, we are working on the problem of threshold-based buffer control, for which IPA can be used to optimally determine the threshold value.
Advisor: Christos Cassandras
Yufan Luo Advisor: John Baillieul
Mohammad Moghadasi
My research interests lie in the field of Computational Biology and Bioinformatics with emphasis on Proteomics, and my current research focuses particularly on Protein-protein docking and Prediction of protein structure. My job involves with different fields of science such as Molecular Biology, Statistical Physics, Molecular Dynamics, and various mathematical tools such as Nonlinear Optimization. I am working on the development of new, efficient methods to predict and analyze the protein-protein interactions that play a crucial role in various aspects of the structural and functional organization of the cell.
Advisor: Ioannis Paschalidis
Feng Nan
I’m interested in supervised machine learning under budget constraint. As machine learning algorithms become increasingly pervasive in big data analytics, computation and feature acquisition costs begin to play critical roles in real life applications. My research focuses on designing efficient algorithms that reduce such costs. An application area is active sensing, where the aim is to balance classification accuracy and the cost of taking sensor measurements.
Advisor: Ioannis Paschalidis
Elli Ntakou
I am working with Prof. Caramanis on the extension of Locational Marginal Prices to distribution level networks to (i) incorporate marginal costs of real and reactive power, losses, voltage control and distribution asset life degradation and (ii) enable medium and low voltage consumers and distributed generation to provide reactive power compensation, voltage control and loss reduction. Today’s average cost pricing practice deprives millions of consumers from the opportunity to match their preferences to distribution system marginal costs and wastes the opportunity to capture significant cost reducing efficiencies and to assist the cost efficient integration of clean distributed generation. Recent critical developments in communication, computation, the advent of flexible loads and power electronics, advocate a major power market reform. We propose a complete market framework that optimizes costs minus distributed participant utility subject to full AC load flow relations and that includes distributed participant costs and benefits and models degrees of freedom ranging from the ability to delay and reschedule consumption to putting excess power electronics resources to dual use for VAr compensation. Addressed issues include the non-convexities of the AC load flow, computational complexity, the resulting clearing prices and distribution network rent, distributed generation optimal location and income, market size, and distributed ways for market clearing.
Advisor: Michael Caramanis
Francisco Penedo Alvarez Advisor: Calin Belta
Sepideh Pourazarm
I work in the CODES lab under supervision of Prof. Christos Cassandras. My research interest is modeling and  optimization of energy-aware systems. More specifically, I work on optimal routing in Wireless Sensor networks.  We are developing an optimal control approach to solve the  problem of routing in sensor networks where the goal is to minimize the network’s lifetime. In our analysis, the energy sources (batteries)   are assumed ‘non-ideal’ and behaving according to a dynamic energy consumption model which captures the nonlinear behavior of actual batteries. Another problem I am working on is optimal routing of Electric Vehicles with energy constraints. In this problem, we seek to minimize the total elapsed times for EVs to reach their destinations by determining routs as well as recharging amounts when the vehicles do not have  adequate energy for the entire journey.
Advisor: Christos Cassandras
Saeedeh Salimianrizi Advisor: Michael Caramanis
Michael Sangillo Advisor: Ishwar Prakash
Wei Si
I work with the Laboratory of Networking and Information Systems (NISLAB). We are involved in providing novel perspectives to modern networking with emphasis on scalability, heterogeneity, and performance. Our research roots into the mathematical fields of graph theory and algorithms, probability and stochastic processes, and coding theory with applications to security, content synchronization, network monitoring, wireless spectrum management, and advanced networking for scientific applications.
Advisor: David Starobinski
Eran Simhon
I work in the NISLAB on costumers’ behavior analysis of different queueing systems that support advance reservations. Our goal is to have a better understanding of the economic aspects of those systems. In particular, we are interested in finding for which settings, enabling advance reservation will increase the providers profit and/or the social welfare. We focus on  applications in the field of communication networks and cloud computing though our results can be applied to other fields. The tools we are using are Game Theory, Linear and Non-Linear Optimization, Stochastic Process and Dynamic Programming.
Advisor: David Starobinski
Xinmiao Sun
I am working with Professor Cassandras in the CODES Lab and our research interests are coverage control and optimization problems. We seek to use distributed algorithms to control the deployment of sensor nodes to maximize the coverage quality in a mission cluttered with obstacles. Currently, we are focusing on finding efficient algorithms to escape a local optimum in the coverage control problem. We test our algorithm by java simulation now and will do it with real robots in the lab.
Advisor: Christos Cassandras
Rebecca Swaszek Advisor: Christos Cassandras
Cristian-loan Vasile
I work in the Hybrid and Networked Systems Lab with Calin Belta. My research is focused on motion planning for high dimensional robotic systems which must satisfy rich mission specifications. The specifications are given as linear temporal logic formulae and are defined over propositions assigned to regions of interest in a given environment. We approach such problems using sampling based methods, a class of randomized algorithms, which are capable of rapidly exploring high dimensional spaces, and combine them with automata based model checking methods. My goal is to develop a framework for automatic deployment of robotic teams which need to cooperate in order to perform persistent tasks under communication and charging constraints.
Advisor: Calin Belta
Shuai Wang

I work in the Intelligent Mechatronics Lab with Professor John Baillieul on developing bio-inspired autonomous systems. We seek to develop new principles for perceptual-based control of autonomous air vehicles around complex and unstructured terrains based on studies of animal flight behavior. More specifically, we focus on understanding of the flight behavior of bats, especially how they navigate inside their habitat and how they integrate heterogeneous sensory information. The tools we are using are control theory, linear and non-linear optimization, stochastic process and digital signal processing.
Advisor: John Baillieul
Taiyao Wang Advisor: Christos Cassandras
Liangxiao Xin Advisor: David Starobinski
Tingting Xu Advisor: Ioannis Paschalidis
Fatma Yanikara Advisor: Michael Caramanis
Bowen Zhang
My research lies in the area of smart grid. I explore the possibility to schedule electricity consumption for both residential and commercial buildings where smart appliances have been adopted. Thermostatic appliances is open to demand response since the thermostatic points could be adjusted to electricity prices and temperature could be controlled within comfortable zone with predetermined policies. The smart grid research is correlated with multi-disciplinary knowledge such as dynamic programming, optimization, queuing system, and control theory. Our research also deals with the system identification of electric consumption pattern in actual buildings. We are going to deploy some current transformers to monitor the electric consumption pattern of 15 St. Mary’s, both steady state electric power flow and transient performance would be studied once the data from current transducers are collected. Aside from the issues discussed above, we are also interested in the planning of reserve selling, peak demand consumption shifting, etc.
Advisor: John Baillieul
Jing Zhang
I am working with Professor Paschalidis. My research lies in two-fold. One is using applied probability, stochastic processes, and optimization theory to deal with modelling and algorithms development. In particular, currently we are working on a problem regarding an anomaly detection algorithm, which aims to improve the accuracy of the threshold needed by the generalized Hoeffding test. To outperform the existing methods (typically using the large deviations theory), we are trying to establish weak convergence results for certain quantities of interest. The other is using machine learning approaches to tackle health care problems, making predictions of the necessity of hospitalization for people in the US.
Advisor: Ioannis Paschalidis
Yue Zhang Advisor: Christos Cassandras
Qi Zhao
I am working with Professor Yannis Paschalidis in the Network Optimization and Control Lab. The first research lies in the health care area. It aims to model the effect of Bivalirudin (a medicine used in cardiac surgical patients in ICU) and then make the automatic dosage decision based on the model and data we have. This project involves nonlinear optimization, machine learning, adaptive control theory and so on. The second research topic lies in the biology area. The objective is to infer the real objective function of bacteria. This is an inverse optimization problem that we need to infer the objective function based on the optimal solution we have.
Advisor: Ioannis Paschalidis
Nan Zhou Advisor: Sean Andersson
Master of Science Students
Marina Almeida Advisor: Michael Caramanis
Allen Huang
Zhiyuan Huang Advisor: James Perkins
Cody Nabong Advisor: Theodore Fritz
Zixuan Pan Advisor: Mac Schwager
Wanqing Yang Advisor: Ioannis Paschalidis
Master of Engineering Students
Lan He Advisor: Hua Wang
Yingyu Jiang Advisor: James Perkins
Minxing Li Advisor: Sean Andersson
Dery Partoni
Jolly Pradhan Advisor: Mac Schwager
Michael Schmidt Advisor: Sean Andersson
Pichaya Tingthanathikul
Siqi Wu Advisor: Mac Schwager
LEAP Students
Michael Cicerone Advisor: James Perkins
Jeffrey Hung Advisor: James Perkins
Lauren Kiefer Advisor: James Perkins
Trent Montgomery Advisor: James Perkins
Abhinav Navani Advisor: James Perkins
Minor Students
Alejandro Eguren Advisor: Kamal Sen
Tru Hoang Advisor: Victor Yachot
Patrick Crawford Advisor: Ayse Coskun
Ada Wong Advisor: Prakahs Ishwar