*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Ph. D. Thesis Proposal
By
Johnny L. Worthy III
(Advisor: Prof. Marcus J. Holzinger)
3:00 PM, Friday, December 11, 2015
Montgomery Knight Building
Room 317
Title:
Initialization of Bayesian State Estimation using Partial Information in the Presence of Systemic Uncertainty
Abstract:
Space situational awareness (SSA) and space domain awareness are both national and global priorities as there is an ever-increasing human presence in Earth orbit. The current priority for the United States Joint Space Operations Center is to fully characterize the space environment in Earth orbit and there is network of optical and radar systems used to support this mission. Optical and radar systems are limited primarily by their inability to fully observe the state of space objects, and there is significant need to improve the ability to perform state estimation with only partial information.
This thesis advances the state of the art of state estimation in these systems through the use of uninformative priors. Uninformative priors, specifically the admissible region as applied to observation of space objects, have typically been used to generate a uniform prior distribution from which to initiate sequential state estimators. A framework for the admissible region is shown that analytically accounts for systemic uncertainties in the system, improving state estimation by better representing the prior as a non-uniform distribution. The resulting non-uniform distribution can then be used to determine the probability that two uncorrelated observations are associated. Further, it is shown that if two non-uniform distributions are associated, then the intersection region represents a probability distribution about the state estimate. The non-uniform uninformative prior and the probability of association concepts introduced in this work are not only directly applicable to the optical and radar systems used for SSA, but also are applicable to a wide range of observer architectures. These concepts are applied to short-baseline time difference of arrival observer architectures to demonstrate a novel application of state estimation using partial information.
Committee Members:
Dr. Marcus Holzinger (advisor)
Dr. Robert Braun
Dr. Glenn Lightsey
Dr. Chis Barnes
Dr. Daniel Scheeres
Dr. Travis Blake (Lt. Col., Ret.. USAF)
