Handle |
Handle
http://hdl.handle.net/11134/20002:860653032
|
Persons |
|
Title |
|
Origin Information |
|
Parent Item |
|
Resource Type |
|
Digital Origin |
|
Description |
Description
Tracking algorithms are used in many applications to provide estimates of states (position, velocity, etc.) of targets from noisy measurements. These estimates can be used for predicting future target states. Some possible targets that may be of interest (and that we will consider here) include aircraft, ships, and missiles. This dissertation looks at several real-world scenarios and develops new tracking algorithms to accurately and efficiently solve these problems. These algorithms are compared to the current state-of-the-art and shown to be superior in position and velocity RMSE, or in computational complexity. We investigate three real-world tracking scenarios. First, we develop a new algorithm with low computational complexity for tracking closely spaced targets. Second, we apply a regularized particle filter to the banana and contact lens problems using a multidimensional version of the Epanechnikov kernel for state vectors, developed in the course of the research for this dissertation. Finally, we develop a generalization of the ML-PMHT and apply it to several Over-the- Horizon radar scenarios.
|
Genre |
|
Organizations |
|
Held By |
|
Rights Statement |
|
Use and Reproduction |
Use and Reproduction
These materials are provided for educational and research purposes only.
|
Local Identifier |
|