|
The following research projects are currently in need of dedicated Postdoctoral Fellows to complete their work. To be a
candidate for any of the projects, you must be a U.S. citizen and have received your PhD by the fellowship start date. These
listings are for information only. Applications must be made to the University and Principal Investigator directly.
| Status: FILLED |
Added: 08/27/2008 12:00 AM |
| Research Area: In-Space Robotic Assembly (IRSA) |
Project Description:
The purpose of this research is to develop a methodology for hardware-in-the-loop (HIL) assessment of in-space robotic assembly of a telescope in ?-g in a more cost-effective and risk-tolerant manner using the SPHERES facilities at the MIT SSL and aboard the International Space Station. For ISRA, these tests will need a modular, segmented telescope that can phase its primary mirror to optical tolerances to compensate for misalignment. Docking ports, deployable booms, and manipulators need to be developed. Three concepts for on-orbit assembly of space telescopes are to be studied: hexagonal rings of segment mirrors using dexterous manipulators; solar-sail technology and collective robotic path planning to self-assemble a large reflector; and autonomous deployment using a robotic arm to remove stowed segments and position them. This work also presents some concepts for on-orbit servicing. An Implementation Plan will describe the design, fabrication, cost, schedule, test objectives and test plan for testing ISRA with phase-able optics in the three-step sequence as well as document all of the technical and programmatic information needed to make an informed decision of whether or not to proceed with such a sequential approach in year three.
Requirements:
A U.S. citizen who has recently graduated or will be graduating in December 2008 and is interested in working as a Postdoctoral Fellow at MIT. A PhD in Aero/Space Engineering (Control Systems area preferred) or a similar field (Mechanical, Electrical engineering). Experience with hardware-in-the-loop testing a plus. Work/study of telescopes / space telescopes also a plus.
Application material should include an introduction letter, full curriculum vitae, and the names of three references, including their e-mail addresses and telephone numbers.
This is a two year appointment with the possibility of extension to a third year. Expected date is Fall/Winter 2008.
If you are interested in joining this research project as a Postdoctoral Fellow, please contact:
Dr. David Miller
MillerD@MIT.edu
http://web.mit.edu/aeroastro/people/miller.html
Further information about the position is available from Sharon Leah Brown slbrown@mit.edu.
|
| Status: FILLED |
Added: 08/19/2008 12:00 AM |
| Research Area: Statistical Classification with a Large Number of Highly Dependent Variables and a Large Number of Classes |
Project Description:
New statistically-sound methods are sought that can effectively address the limitations of biometric identification involving large numbers of highly dependent variables.
Technical Objectives: In discrimination problems with a large number of potential classes it is reasonable to expect that the training set does not contain observations from all classes in the population. Most types of biometric identification systems are real world examples of discrimination problems with a large number of classes. The problems related to biometric identification often involve large numbers of highly dependent variables. Currently, available techniques do not fully address these issues, especially with large numbers of classes.
New statistically-sound methods are sought that can effectively address these limitations. These methods may or may not be related to classical methods for classification.
Requirements:
A U.S. citizen who has recently graduated with a PhD in Statistics or a related field or will be graduating by December 2008 and is interested in working as a Postdoctoral Fellow at George Mason University. The Postdoctoral Research Fellow should have a thorough knowledge of modern statistical methods and statistical computing.
If you are interested in joining this research project as a Postdoctoral Fellow, please contact:
Dr. Donald Gantz
703-993-1695
dgantz@gmu.edu
http://statistics.gmu.edu/pages/forensicslab.html
|
| Status: OPEN |
Added: 08/11/2008 12:00 AM |
| Research Area: Image-Based Determination of Polarized Bidirectional Reflectance Distribution Function for In-Field Characterization of Materials |
Project Description:
The goal of this research is to improve technical intelligence capabilities to characterize land-, air- and space-based objects through polarimetric measurements. The polarization dependence of light scattering from object surfaces provides significant information about the nature of that surface, such as material properties, roughness, etc., particularly when all polarization states of the incident and scattered light are considered. If a Stokes vector is used to represent the incident and reflected (or scattered), polarization states, then the transformation matrix, or Mueller matrix, is a characteristic of that surface. Gonio-reflectometer methods traditionally used to measure bidirectional reflectivity distribution functions (BRDF’s) also are used to determine the Mueller matrices associated with scattered light. Image-based BRDF-measurement techniques have recently been developed by the computer-graphics community, and are now being applied by the science community to physical measurements. The inclusion of Mueller-matrix determination in image-based BRDF measurement techniques is a novel application that could significantly increase the information available to polarimetric-classification databases. BRDF values can be difficult, if not impossible, to obtain in foreign or denied territory; this research should also include the development of scene-based BRDF-measurement techniques and the extraction of polarimetric information from them.
Technical Objectives:
• Develop techniques and methodologies for measurement and validation of image-based BRDF’s.
• Develop techniques and methodologies for measurement and validation of Mueller matrices from image-based BRDF data.
• Evolve the image-based BRDF technique such that measurements could be applied to non-cooperative targets in the field.
• Develop physics-based, i.e., truly predictive rather than merely representative as most current BRDF models are, BRDF models.
Requirements:
A U.S. citizen who has recently graduated or will be graduating in December 2008 and interested in working as a Postdoctoral Fellow at Rochester Institute of Technology.
If you are interested in joining this research project as a Postdoctoral Fellow, please contact:
Dr. Carl Salvaggio
585-475-6380
salvaggio@cis.rit.edu
http://www.cis.rit.edu/~cnspci/
|
| Status: OPEN |
Added: 08/08/2008 12:00 AM |
| Research Area: Continuous- Wave Atom Lasers: A Chip-Based Approach |
Project Description:
This topic will investigate the development of micro fabrication techniques for laser cooling and trapping neutral atoms on an atom chip. The research might combine state-of-the-art techniques in micro fabrication, quantum physics, and atomic-optical science.
Technical Objectives:
Experiments could include fabrication and analysis of atom chips and characterization of these devices for quantum information and metrology applications. A second generation atom chip will require computer modeling and optimization for atom chips. Investigations of quantum feedback for ensemble based quantum information processing will be pursued.
Requirements:
A U.S. citizen who has recently graduated or will be graduating in December 2008 and is interested in working as a Postdoctoral Fellow at the University of Michigan.
If you are interested in joining this research project as a Postdoctoral Fellow, please contact:
Dr. Georg Raithel
734-647-9031
graithel@umich.edu
http://www.physics.lsa.umich.edu/department/directory/bio.asp?ID=264
|
| Status: OPEN |
Added: 08/08/2008 12:00 AM |
| Research Area: Mathematical Programming Model for Social Network Role Identification |
Project Description:
Large data sets often are represented as a graph, where nodes represent entities and edges represent the relationship between them. This representation is particularly effective when the data is transactional (e.g. e-mails), but with extraction techniques, other data such as free form text can produce graphs as well. Social Network Analysis (SNA) has been very effective in analyzing graphs. These techniques have been useful in finding groups, leaders within groups and informal networks that operate independently from an organization’s formal structure.
Social Network techniques have been very effective in situations where the graph is from a single source and relatively complete. What is needed to make SNA broadly effective is to develop techniques that take into account content, uncertainty, incompleteness, and heterogeneity of data sources. For large data sets this analysis must be done with minimal human intervention.
Technical Objectives:
The goal of this research topic is to improve the results and applicability of social network analysis by exploring and developing fundamental extensions to the basic graph analysis techniques. Research should be demonstrated on real or realistic data. The topic description above outlines a broad area of research and it is not expected that the research will cover all aspects of the problem. Listed below are some sub-topics:
• Determining the strength of a relationship between two identities using content and multiple relationships
• Characterization of identified groups and cliques
• Using content to describe association of an entity to multiple groups
• Developing techniques that account for uncertainty of entity identification and relationships that may arise from entity extraction.
• The research will focus primarily on the combination of automated text analysis techniques with network programming (i.e., optimization) to identify group substructures of interest.
Requirements:
A U.S. citizen who has recently graduated or will be graduating in December 2008 and is interested in working as a Postdoctoral Fellow at Oak Ridge National Laboratory.
The ideal candidate will have experience with linear and non-linear optimization (modeling and solution in large-scale systems) or automated text analysis methods (Latent Semantic Indexing, Latent Dirichlet Analysis etc.). Research will focus on the applications of these methods to the deep analysis of social networks, social pattern discovery and automated role detection theory.
If you are interested in joining this research project as a Postdoctoral Fellow, please contact:
Dr. Christopher Griffin
865-574-9389
griffinch@ornl.gov
http://orise.orau.gov/sep/needs/files/ORNL08-103-CSED.pdf
|
|
| |
 |
| Home : Privacy Policy : Accessibility : Contacts |
U.S. Government Proprietary - All Rights Reserved |
|
