Advancements of Particle Filtering Theory and it's Application to Tracking
Abstract
The main goal of this work is the development of a general class of particle filtering methods and apply them to various problems related to target tracking. Theoretically, the project involves the advancement of existing particle filtering schemes and the development of new ones that relax the probabilistic assumptions of the standard methods. The design of the new filters is guided by the objectives of securing (a) excellent performance in target tracking in most demanding situations, (b) robustness, and (c) relatively easy hardware implementation. Practical efforts include applications of the filters to tracking of single targets as well as to much more challenging tasks such as tracking of multiple targets where the number of targets may vary with time. Finally, scenarios that require multisensor tracking and data fusion are also of interest.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2006
- Accession Number
- ADA458305
Entities
People
- Monica F. Bugallo
- Peter M. Djuric
Organizations
- Stony Brook University