Kalman filter implementation in matlab environment

Kalman filter (KF) is one of the famous recursive algorithm developed in the twentieth century to solve the problem of discrete-data linear filtering. Since it was first introduced in the paper written by R.E. Kalman, it has been widely researched on and implemented in various fields such as automat...

Full description

Saved in:
Bibliographic Details
Main Author: Yeo, Kelvin.
Other Authors: School of Electrical and Electronic Engineering
Format: Final Year Project
Language:English
Published: 2013
Subjects:
Online Access:http://hdl.handle.net/10356/52605
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Kalman filter (KF) is one of the famous recursive algorithm developed in the twentieth century to solve the problem of discrete-data linear filtering. Since it was first introduced in the paper written by R.E. Kalman, it has been widely researched on and implemented in various fields such as automation, navigation and even economics. The capabilities of the algorithm is not limited to just producing estimation from a noisy source, it is also able to estimate the past and future state given the current state. [1] This method comprises of 2 parts in the iteration, namely the prediction step and the update step. The objective of this project is to analyze and understand the algorithm of the Kalman Filter (KF) which will be implemented for sound localization using Matlab. The methods used in this project for sound localization is Time Difference of Arrival (TDOA) and the Levenberg Marquardt (LM) optimization method. The purpose of KF in this system is to improve the results produced by the LM, thus making the system more robust. The project is carried out in 3 stages. The first stage is to understand the KF algorithm using Matlab. The second stage would be the testing of the KF algorithm in a closed environment to determine the capability of the algorithm together with the TDOA and LM algorithms. After which, the codes will be modified and transported to the real-time system for stage 3 testing. It is an opportunity to put the KF algorithm to the test and observe the characteristic of the filter in a real-time system. Through experiment testing, we can see if the KF is sufficient to make the system more robust.