|
|
|
|
Project List |
|
- Professor: Paul Rosenbloom
Department: Computer Science
Website: http://http://cs.usc.edu/~rosenblo/
Description: Creating a new form of cognitive architecture for virtual humans (and intelligent robots/agents) based on graphical models (factor graphs). Work is in progress on memory, decision making, problem solving, learning, perception, localization and mapping, mental imagery, natural language, and theory of mind. Looking for a student interested in developing, applying (to sample tasks), analyzing and/or evaluating new architectural capabilities.
Number of Students: 1
Prerequisites: Interest in cognitive architecture and ability to program
Comments:
- Professor: Peter Beerel
Department: EE Systems - Computer Engr. Grp.
Website:
Description: Despite the numerous asynchronous designs that have demonstrated substantial benefits over their synchronous counterparts, the lack of CAD tools has been a significant hurdles in the path to wide-spread adoption of asynchronous logic. Proteus is a new flow developed by the USC start-up company TimeLess Design Automation now in use by Intel that targets 2-3X higher performance than typically possible for synchronous counterparts, namely 1.1GHz in TSMC 65nm. The Proteus flow leverages both synchronous synthesis and place-and-route tools and supports specifications both in standard RTL and SystemVerilog.
Our current research efforts involve reducing the power consumption of the resulting circuits, enabling the use of more standard-cell libraries, and automating the application of the flow to network on chip applications.
Number of Students: 2
Prerequisites: Strong C++ and VLSI knowledge.
Comments: Students will work with a team of graduate students to build sophisticated software tools that help design low-power high-performance digital circuits building on a recently commercialized software platform. They must have strong software background, be motivated to learn new things, and have superlative communication skills.
Given the complexity of this project, students must be able to spend 7-8 weeks on this effort.
- Professor: Yan Liu
Department: Computer Science
Website: http://http://http://http://www-bcf.usc.edu/~liu32/group/melady.html
Description: Social media have been commonly used since the advent of Web 2.0, either for personal use at leisure or official use in an organization,. It has become a ubiquitous tool for people’s daily social communications. Obviously, there is the pressing need of an intelligent solution to help users organize social media as well as the associated tasks for better social collaboration. In this project, we will explore effective machine learning approaches for analyzing social media content, making social recommendation and detecting social anomalies.
Number of Students: 2
Prerequisites: Algorithms, C++/java coding skill
Comments:
- Professor: Qiang Huang
Department: Epstein Dept of Industrial & Systems Engineering
Website: http://www-rcf.usc.edu/~qianghua
Description: Unlike traditional manufacturing, nanomanufacturing (NM) involves product characteristics and process variables at multiple feature scales. However, control of the multiscale process/product variations (MPVs) in NM is constrained by the scarcity of measurement data, a lack of observation during processing, and limited physical knowledge. This project intend to integrate nano informatics with nano manufacturing processes to address the challenge of MPVs for nanotechnology scale-up
Number of Students: 1
Prerequisites: Mechanics, Mathematical physics
Comments:
- Professor: Qiang Huang
Department: Epstein Dept of Industrial & Systems Engineering
Website: http://www-rcf.usc.edu/~qianghua
Description: Unlike traditional manufacturing, nanomanufacturing (NM) involves product characteristics and process variables at multiple feature scales. However, control of the multiscale process/product variations (MPVs) in NM is constrained by the scarcity of measurement data, a lack of observation during processing, and limited physical knowledge. This project intend to integrate nano informatics with nano manufacturing processes to address the challenge of MPVs for nanotechnology scale-up
Number of Students: 2
Prerequisites: Mechanics, Mathematical physics
Comments:
- Professor: Hossein Hashemi
Department: Electrical Engineering/Electrophysics
Website:
Description: Prof. Hashemi research group has been working on chip-scale radars and algorithms to detect, localize, and track humans and identify their features such as gait in complex environments.
Applications of this research include healthcare (smart homes and hospitals) for monitoring the patients, elderly, and children; security; gaming and entertainment; and automation.
Undergraduate research opportunities include contributions to the sensor hardware development, embedded programming, and software development (for algorithms or GUI).
Number of Students: 2
Prerequisites: Top junior or senior in Eelectrical Engineering or Computer Engineering
Comments:
- Professor: Michelle Povinelli
Department: Electrical Engineering/Electrophysics
Website: http://http://usc.edu/nanophotonics
Description: Conduct theoretical research involving optical modeling of nanostructured materials.
Number of Students: 1
Prerequisites: electromagnetism, quantum mechanics
Comments:
- Professor: Milind Tambe
Department: Computer Science
Website: http://http://http://teamcore.usc.edu/projects/security/
Description: Game Theory for Security:
Security is a critical concern around the world that arises in protecting our ports, airports or other critical infrastructure from adversaries, in protecting our wildlife and forests from poachers and smugglers, and in curtailing the illegal flow of weapons, drugs and money. In all these cases, we have limited security resources that must be deployed intelligently taking into account differences in priorities of targets and the responses of the adversaries.
Algorithmic Game theory is well-suited to adversarial reasoning for security resource allocation and scheduling problems. Our newly developed game-theoretic algorithms are now deployed in multiple applications: ARMOR has been deployed at the Los Angeles International Airport (LAX) since 2007 to randomizes checkpoints and canine patrol routes within the airport terminals; IRIS for randomized deployment of the Federal Air Marshals (FAMS) since 2009; PROTECT is deployed in the port of Boston for randomizing US coast guard patrolling ; GUARDS is under evaluation for national deployment by the TSA.
This research project will focus on algorithms to help security agencies in many different real-world domains.
Number of Students: 2
Prerequisites: Probabilities, calculus, programming
Comments:
- Professor: Ellis Meng
Department: Biomedical Engineering
Website: http://biomems.usc.edu
Description: The Biomedical Microsystems Laboratory develops electrochemically-based sensors and actuators that have a variety of biomedical applications from drug delivery to pressure sensing. The lab seeks motivated students interested in conducting research on device fundamentals including transduction mechanisms as well as systems integration efforts towards wireless implantable microdevices. Students having background in wireless circuit, mechanical component, microfluidic system design and testing are highly encouraged to apply. Students are expected to read and understand research literature, design experimental setups, collect data, write technical reports, and present their results during the internship.
Number of Students: 2
Prerequisites: junior-level standing
Comments:
- Professor: Anupam Madhukar
Department: Materials Science
Website: http://nanostructure.usc.edu
Description: Quantum nanostructures are the central players in device structures underlying such applications as solar energy conversion and quantum computing. This project involves(1)synthesis of quantun dots via colloidal solution growth or vapor phase deposition on naotemplated substrate,(2)characterization using photoluminescence,&(3)calculations to analyze the data.
Number of Students: 2
Prerequisites: Junior level engineering math, basic physics and chemistry, modern / quantum physics,
Comments: A one semester introductory course in quantum mechanics is helpful background for project oriented towards either solar energy conversion or quantum computing.
- Professor: Shrikanth Narayanan
Department: Signal & Image Processing
Website: http://http://sail.usc.edu
Description: Participate and contribute to research and development in fundamental and applied research in human-centered information processing
at the Signal Analysis and Interpretation Laboratory (SAIL).
Our emphasis is on speech, audio, language, biomedical and multi-modal signal processing, machine learning and pattern recognition.
SAIL's research applications and systems development especially focus on domains with direct societal relevance including in human health and well being, education and defense.
SAIL supports a collaborative interdisciplinary environment and bridges research from several departments and schools both within and outside USC.
Number of Students: 2
Prerequisites: Please see comments below.
Comments: Requirements: Strong interests in one or more of the
following: Signal Processing, Computing, Machine Learning, Speech
science, Human communication, Natural Language Processing, Affective
Computing. Programming skills. • Additionally a desire to work on real problems and data, to take leadership
and be proactive, ability to work in a collaborative manner, and a thirst for
learning.
- Professor: Murali Annavaram
Department: EE Systems - Computer Engr. Grp.
Website: http://http://www.usc.edu/dept/ee/scip/research/
Description: This project is focused on benchmarking intermittent errors on hardware. We will look at how instructions when executed on hardware can age the processor and cause the processor to fail.
Number of Students: 1
Prerequisites: Good C coding and RTL skills
Comments:
- Professor: Urbashi Mitra
Department: Communication Sciences Institute
Website: http://knowme.usc.edu
Description: We are designing methods and algorithms for wireless body area sensing networks. We wish to develop power allocation and signaling schemes for such wireless systems as well as signal processing methods for feature extraction, physical state detection and state tracking.
Number of Students: 1
Prerequisites: probability, linear algebra, basic programming
Comments: Students will work with real sensors, but will also be required to be independent in scanning the biomedical engineering literature as well as that in more classical electrical engineering and computer science.
- Professor: Urbashi Mitra
Department: Communication Sciences Institute
Website:
Description: We are designing signal processing algorithms for anomaly detection in large scale networks such as the SmartGrid, commercial cellular networks, etc. using models and new wavelet families.
Number of Students: 1
Prerequisites: probability, linear algebra, basic programming, Markov chains, networking
Comments: This will be a mostly mathematical, algorithmic derivation project with simulation to verify the theory.
- Professor: Urbashi Mitra
Department: Communication Sciences Institute
Website:
Description: We are exploiting new models for wideband channels such as underwater acoustic and ultrawideband radio. The goal is to develop novel equalizers for such complex, time-varying channels.
Number of Students: 1
Prerequisites: signal processing, linear algebra, probability theory, communications
Comments:
- Professor: Shuo-Wei Chen
Department: Electrical Engineering/Electrophysics
Website: http://http://www-bcf.usc.edu/~swchen/index.htm
Description: The research project is to investigate the analog and digital signal processing techniques for high-performance, low-power mixed signal circuit designs, such as analog to digital converter (ADC) and digital to analog converter (DAC). The research involves algorithmic exploration, behavioral modeling, transistor-level design, PCB board design and lab testing. It provides the student great opportunities to experience the state of the art circuit design techniques and exposure to multi-domain knowledge in both analytical and practical perspectives.
Number of Students: 1
Prerequisites: analog and digital VLSI courses
Comments:
- Professor: Shuo-Wei Chen
Department: Electrical Engineering/Electrophysics
Website: http://http://www-bcf.usc.edu/~swchen/index.htm
Description: The research project is to investigate the analog and digital signal processing techniques for high-performance, low-power mixed signal circuit designs, such as analog to digital converter (ADC) and digital to analog converter (DAC). The research involves algorithmic exploration, behavioral modeling, transistor-level design, PCB board design and lab testing. It provides the student great opportunities to experience the state of the art circuit design techniques and exposure to multi-domain knowledge in both analytical and practical perspectives.
Number of Students: 1
Prerequisites: analog and digital VLSI courses
Comments: N/A
- Professor: Shuo-Wei Chen
Department: Electrical Engineering/Electrophysics
Website: http://http://www-bcf.usc.edu/~swchen/index.htm
Description: The research project is to investigate the analog and digital signal processing techniques for high-performance, low-power mixed signal circuit designs, such as analog to digital converter (ADC) and digital to analog converter (DAC). The research involves algorithmic exploration, behavioral modeling, transistor-level design, PCB board design and lab testing. It provides the student great opportunities to experience the state of the art circuit design techniques and exposure to multi-domain knowledge in both analytical and practical perspectives.
Number of Students: 1
Prerequisites: analog and digital VLSI courses
Comments: N/A
- Professor: Gerard Medioni
Department: Computer Science
Website: https://docs.google.com/View?id=df4bwzm8_26mh9tbndt
Description: Given an input video, generate accurate 3D model of the face.
We will consider input from a 3D camera (Kinect)or a regular webcam
Number of Students: 2
Prerequisites: C++, Matlab programming
Comments:
- Professor: Gerard Medioni
Department: Computer Science
Website: http://none
Description: Build 3D model of a person from different views generated by a 3D camera (Kinect/PrimeSense)
Number of Students: 2
Prerequisites: C++, Matlab programming
Comments:
- Professor: Gerard Medioni
Department: Computer Science
Website: http://none
Description: Build 3D model of a person from different views generated by a 3D camera (Kinect/PrimeSense)
Number of Students: 2
Prerequisites: C++, Matlab programming
Comments:
- Professor: Gerard Medioni
Department: Computer Science
Website: http://none
Description: Given a geo-map, and a detected event from video, display a geo-registered visualization of the event from an arbitrary view point.
Number of Students: 1
Prerequisites: C++, Matlab programming, Graphics
Comments:
- Professor: Ulrich Neumann
Department: Computer Science
Website: http://graphics.usc.edu
Description: Our overall goal is to automatically construct 3D models from sensor data (3D point clouds and image data). Several PhD students are actively working on aspects of this problem. See the recent papers on the lab web site for a better understanding of what we do.
Number of Students: 1
Prerequisites: see related background
Comments: Work independently or join the group efforts. We can accommodate you at many levels depending on your interests and experience.
- Professor: Paul Ronney
Department: Aerospace and Mechanical Engineering
Website: http://http://ronney.usc.edu/Research/MicroFIRE/
Description: It is well known that the use of combustion processes for electrical power generation provides enormous advantages over batteries in terms of energy storage per unit mass and in terms of power generation per unit volume, even when the conversion efficiency in the combustion process from thermal energy to electrical energy is taken into account. For example, hydrocarbon fuels provide an energy storage density between 40 and 50 MJ/kg, whereas even modern lithium ion batteries commonly used in laptop computers provide only 0.4 MJ/kg. Thus, even at only 5% conversion efficiency from thermal to electrical energy, hydrocarbon fuels provide about 5 times higher energy storage density than batteries.
We are studying several means to generate electrical power at small scales including thermoelectric devices, many types of fuel cells, thermal transpiration pumping for propulsion, etc. Potential interns capable of executing a clear, focused, logical, "think outside the box" approach are invited to apply.
Number of Students: 1
Prerequisites: Understanding of experimental methods, thermodynamics, fluid mechanics
Comments:
- Professor: Paul Ronney
Department: Aerospace and Mechanical Engineering
Website: http://http://ronney.usc.edu/research/EdgeFlames/EdgeFlames.html
Description: Flames subject to temporally and spatially uniform hydrodynamic strain are frequently used to model the local interactions of flame fronts with turbulent flow fields. The "laminar flamelet" concept presumes that each surface element of the flame front behaves as though it were a steady isolated front subject to uniform strain. The applicability of laminar flamelet models in strongly turbulent flows is questionable because in turbulent flows the strain rate (s) changes at rates comparable to s itself and the scale over which the flame front curvature and s changes is comparable to the curvature scale itself.
As a step towards more realistic quantification of strain effects in turbulent premixed flames, we are studying flame structures that occur in the transition region between the extinguished and burning regions of the flame front with the intent of providing better models of turbulent combustion as applied to automotive engines and gas turbines. Summer interns would conduct experimental studies of the effect of strain rate, heat losses, mixture type and strength, etc. on the propagation and extinction of these edge flames.
Number of Students: 1
Prerequisites: Thermodynamics, fluid mechanics, competence in experimental methods
Comments:
- Professor: Paul Ronney
Department: Aerospace and Mechanical Engineering
Website: http://http://ronney.usc.edu/research/biophysics/index.html
Description: While enormous effort has been spent on engineering of conventional fuel cells (CFCs) using hydrogen and methanol, relatively little engineering has been devoted to Microbial Fuel Cells (MFCs). Engineering of any system requires sufficient understanding of its characteristics to develop a predictive model. Development of a model for MFCs is not a simple extension of models of CFCs because of the many differences between MFCs and CFCs. These differences include much lower anode (fuel-side) activity (expressed, for example, in terms of current density); life-support and health requirements of the anode "catalyst" (bacteria); variability and adaptation (both desirable and undesirable) of the anode; and fuel (nutrient) complexity and flexibility. Despite these differences, we believe reaction-diffusion modeling, typically employed in chemical systems such as CFCs and combustion, will be a valuable tool for MFC engineering, especially when combined with genetic approaches, examples of which are described. Furthermore, because of the differences between MFCs and CFCs, it is considered very likely that viable MFCs will require novel architectures not employed in any CFCs.
We are studying candidate architectures for MFCs. Interns would conduct experiments to assess the effect of anode type and geometry, nutrient medium, oxygen supply, etc. to the fuel cell power output and longevity.
Number of Students: 1
Prerequisites: Competence in experimental methods, knowledge of fluid mechanics
Comments:
- Professor: Bhaskar Krishnamachari
Department: Electrical Engineering Systems
Website: http://anrg.usc.edu
Description: Design and Analysis of Algorithms for wireless Networks
Advisor: Bhaskar Krishnamachari, http://ceng.usc.edu/~bkrishna
The Autonomous Networks Research Group (ANRG) seeks bright undergraduate students with backgrounds in electrical engineering, computer science and mathematics for research into next-generation wireless networks (including sensor networks, vehicular networks, underwater networks, green cellular networks, cognitive radio networks, and mobile social networks). The available projects involve a mix of mathematical analysis, simulation and hands-on experiments, and will provide a strong experience in graduate-level research. Former summer interns with this group have gone on to top graduate programs in EE/CS including at Stanford, MIT, UIUC, Princeton, Columbia, U. Michigan, besides USC itself.
Number of Students: 2
Prerequisites: 3 years of undergrad in EE/CS/Mathematics
Comments:
- Professor: Rahul Jain
Department: EE Systems - Computer Engr. Grp.
Website: http://www-rcf.usc.edu/~rahuljai
Description: Electricity markets for the smart grid: The structure and operation of electricity markets in the US and other countries will be investigated. Demands on such markets due to renewable energy integration, smarter grid, and smart meters for demand response will be determined. Some market architectures shall be proposed and analyzed.
Number of Students: 1
Prerequisites: None
Comments: Potential candidates: 1) Dhruv Mishra, IIT Delhi, 2) Akshay Agrawal, IIT Kanpur
- Professor: Rahul Jain
Department: EE Systems - Computer Engr. Grp.
Website: http://www-rcf.usc.edu/~rahuljai
Description: Dynamic Spectrum Management: Challenges and Potential Solutions: We will investigate spectrum management challenges on various wireless networks - cellular, femto-cell, 802.11, cognitive and multi-hop. Various dynamic spectrum management approaches shall be investigated, and challenges identified including due to incentive issues. Some solutions shall be proposed.
Number of Students: 1
Prerequisites: None
Comments: Potential candidates: 1) Dhruv Mishra, IIT Delhi, 2) Akshay Agrawal, IIT Kanpur
- Professor: Maja Mataric
Department: Computer Science
Website: http://http://http://http://robotics.usc.edu/interaction
Description: While there exist a variety of approaches for autonomous learning in
real world domains, few of these algorithms are appropriate for use
with agents which must interact with humans. As part of this project,
the student will help investigate the use of learning algorithms in
human-robot interaction. In particular, this project will focus on
the use of reinforcement learning approaches over probabilistic models in human-centric domains. This will involve development of
appropriate algorithms to account for humans, implement an experiment for data collection, and help conduct a validation with human participants. Implementation will potentially include recognition of human gestures/behavior, robot control for a domain, and implementation or modification of a learning algorithm.
Number of Students: 1
Prerequisites: C/MATLAB Programming
Comments:
- Professor: Amy Rechenmacher
Department: Civil/Environmental Engineering
Website: http://cee.usc.edu/faculty-staff/faculty-directory/rechenmacher-amy.htm
Description: Digital Image Correlation (DIC) is used to quantitatively track and characterize the evolution and fate of vortex structures within localized slip planes that form within dense granular materials undergoing monotonic shear. The student will perform quantitative comparisons among several competing image-based analysis software packages, with the aim of achieving optimal accuracy in grain-scale displacement measurement. Student will also assist in performing some granular shear experiments.
Number of Students: 1
Prerequisites: Mechanics, Soil Mechanics/Geotechnical Engineering courses, programming in Matlab
Comments: Interest and aptitude in experimental work is helpful
- Professor: Sven Koenig
Department: Computer Science
Website:
Description: Auction-based coordination systems coordinate
teams of agents, such as teams of mobile robots
that have to repeatedly reassign targets among
themselves as they learn more about the terrain,
as robots fail or as additional targets get
introduced. Applications include environmental
clean-up, mine clearing, space exploration and
search and rescue. Artificial intelligence and
robotics researchers have explored auction-based
coordination systems at least since the
introduction of contract networks, but mostly from
an experimental perspective. Auctions in economics
deal with competitive agents that often have long
decision cycles, while auction-based coordination
systems deal with cooperative agents that often
have to operate in real-time, which raises a
completely different set of issues. This research
project will involve both theoretical and
experimental approaches for auction-based
coordination systems, to understand their behavior
better and to extend their capabilities.
Number of Students: 1
Prerequisites: ---
Comments:
- Professor: Sven Koenig
Department: Computer Science
Website:
Description: Ridesharing has been viewed as an alternative
transportation means to save travel cost, to
reduce travel time, to mitigate traffic
congestion, to conserve fuel, and to reduce air
pollution. There have been some attempts to create
online ridesharing systems in industry and
academia but never on a large scale, in real-time,
with a sophisticated pricing mechanism, and for a
potentially huge population of registered users to
provide concierge-like ride-matching
services. This project creates a new dynamic
ridesharing market in theory and in
application. We envision dynamic ridesharing
markets that are capable of instantaneous ride
matching with a pricing mechanism that encourages
drivers to sell their unused vehicle seats to
other travelers and incentivize passengers to
participate in the ridesharing community as soon
as possible. The challenge of this project is to
develop new approaches to mechanism design,
multi-agent systems, distributed optimization,
machine learning, computation of large scale
equilibria, planning under uncertainty, and user
interfaces.
Number of Students: 1
Prerequisites: ---
Comments:
- Professor: Jerry Loeb
Department: Medical Device Development Facility
Website: http://http://mddf.usc.edu
Description: We have equipped a Barrett research robotic arm and hand with BioTac multimodal tactile sensors (originally developed in our lab and now available from www.SynTouchLLC.com). We are developing algorithms for exploratory movements and signal processing so that the robot can characterize and identify objects in its environment, similarly to humans.
Number of Students: 2
Prerequisites: computer programming
Comments:
- Professor: Jihie Kim
Department: Information Sciences Institute
Website: http://http://ai.isi.edu/pedtek/pedtek.html
Description: Social Dialogue Analysis: The students will develop social network and dialogue models for
online chat data. The work will focus on question and answer patterns, analyzing how
questions are answered through online dialogue.
Number of Students: 2
Prerequisites: Computer Science
Comments:
- Professor: Yu-Han Chang
Department: Information Sciences Institute
Website: http://http://cb.isi.edu/projects.html
Description: New technologies deployed in the NBA enable automatic collection of fine-grained spatio-temporal data that tracks player and ball movements, opening up a wide range of pattern recognition and sports analysis problems. We develop and apply machine learning, data analysis, and visualization tools to both validate long-held beliefs and uncover the hidden truths behind the game.
Number of Students: 1
Prerequisites: Java
Comments:
|
|
 |
