PARTICLE FILTER APPROACH TO OVERCOME MULTIPATH PROPAGATION ERROR IN SLAM INDOOR APPLICATIONS.

Search for this publication on Google Scholar
Shetty, N. (2018). PARTICLE FILTER APPROACH TO OVERCOME MULTIPATH PROPAGATION ERROR IN SLAM INDOOR APPLICATIONS. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

74 views ◎
39 downloads ⇓


Abstract

Robot localization faces the classic chicken or egg problem where it has to either know the location of the robot or the map to localize itself. In most applications this can be using Simultaneous Localization and Mapping(SLAM) through measurements via sensors. Most applications of SLAM involve taking measurements that contain both range and orientation information about its environment. But Range-Only SLAM localization faces an ambiguity because it receives only range measurements from its sensors. Hence in order to implement localization we would need to know atleast an approximate initial location of the landmarks. Previous approaches to this issue involves basic trilateration, probabilistic methods, least squares approximation, multiple hypothesis methods using the Extended Kalman Filter. This however does provide a delay the localization. In order to overcome this there are multiple hypothesis methods that initialize the problem in real time. However, this is computationally expensive and all of these methods do not cope well with multipath noise. Multipath is the propagation phenomenon that results in radio signals reaching the receiving antenna by two or more paths leading to false range measurements.In this research we explore the problem put forward with the use of radio beacons when applied to RO SLAM and multipath noise that has a non Gaussian distribution. In particular, the use of the particle filter approach that copes with multipath noise as compared to the Extended Kalman Filter that is only applicable to noise that has a Gaussian distribution. This thesis proposes a method using particle filters to perform SLAM and is then compared to the established methods through simulations.

Details
Author

Shetty, Nelyadi

Title

PARTICLE FILTER APPROACH TO OVERCOME MULTIPATH PROPAGATION ERROR IN SLAM INDOOR APPLICATIONS.

Physical Description

1 online resource (66 pages) : PDF

Date

2018

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Robot localization faces the classic chicken or egg problem where it has to either know the location of the robot or the map to localize itself. In most applications this can be using Simultaneous Localization and Mapping(SLAM) through measurements via sensors. Most applications of SLAM involve taking measurements that contain both range and orientation information about its environment. But Range-Only SLAM localization faces an ambiguity because it receives only range measurements from its sensors. Hence in order to implement localization we would need to know atleast an approximate initial location of the landmarks. Previous approaches to this issue involves basic trilateration, probabilistic methods, least squares approximation, multiple hypothesis methods using the Extended Kalman Filter. This however does provide a delay the localization. In order to overcome this there are multiple hypothesis methods that initialize the problem in real time. However, this is computationally expensive and all of these methods do not cope well with multipath noise. Multipath is the propagation phenomenon that results in radio signals reaching the receiving antenna by two or more paths leading to false range measurements.In this research we explore the problem put forward with the use of radio beacons when applied to RO SLAM and multipath noise that has a non Gaussian distribution. In particular, the use of the particle filter approach that copes with multipath noise as compared to the Extended Kalman Filter that is only applicable to noise that has a Gaussian distribution. This thesis proposes a method using particle filters to perform SLAM and is then compared to the established methods through simulations.

Genre

masters theses

Subjects--Topics

Electrical engineering

Degree

M.S.

Subject Area

Electrical Engineering

Advisor(s)

Conrad, James

Committee Members

Willis, Andrew
Weldon, Tom

Degree Note

Thesis (M.S.)--University of North Carolina at Charlotte, 2018.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Shetty_uncc_0694N_11780

Permalink

http://hdl.handle.net/20.500.13093/etd:908