A Breadcrumb Network for Assisting with Autonomous Robot Localization

Search for this publication on Google Scholar
Grabowsky, D. (2021). A Breadcrumb Network for Assisting with Autonomous Robot Localization. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

75 views ◎
23 downloads ⇓


Abstract

Localization and communication are critical components for functioning autonomous robots. The infrastructure required or these operations commonly includes global positioning system (GPS) and easily recognizable and re-identifiable landmarks. However, these types of infrastructures are not always readily available. GPS typically uses a low power signal that can be denied intentionally or is unable to penetrate certain materials. Unique landmarks can be difficult to find in unstructured environments like forests, where many potential landmarks can seem nearly identical. To solve this, this research has developed a deploy-able electronic way-point system dubbed ’BreadCrumbs’. BreadCrumbs function as electronic landmarks that can provide localization and communication capabilities to a robot in environments where such infrastructure is not present. When deployed by a forward moving agent with a set destination, the BreadCrumbs also form a series of way-points which reduce the possible state space an autonomous robot must search through when path planning in an unknown or un-mapped environment. The BreadCrumbs are self localizing and have several methods for initial location determination based on the environment they are placed in. GPS is not required for the BreadCrumbs to function and, once established, they can function as landmarks for autonomous robots by providing range data from radio signal strength with a path loss exponent determined through a Deep Deterministic Policy Gradient algorithm. The algorithm is designed such that a path loss exponent for each BreadCrumb location is learned during run time.

Details
Author

Grabowsky, David

Title

A Breadcrumb Network for Assisting with Autonomous Robot Localization

Physical Description

1 online resource (157 pages) : PDF

Date

2021

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Localization and communication are critical components for functioning autonomous robots. The infrastructure required or these operations commonly includes global positioning system (GPS) and easily recognizable and re-identifiable landmarks. However, these types of infrastructures are not always readily available. GPS typically uses a low power signal that can be denied intentionally or is unable to penetrate certain materials. Unique landmarks can be difficult to find in unstructured environments like forests, where many potential landmarks can seem nearly identical. To solve this, this research has developed a deploy-able electronic way-point system dubbed ’BreadCrumbs’. BreadCrumbs function as electronic landmarks that can provide localization and communication capabilities to a robot in environments where such infrastructure is not present. When deployed by a forward moving agent with a set destination, the BreadCrumbs also form a series of way-points which reduce the possible state space an autonomous robot must search through when path planning in an unknown or un-mapped environment. The BreadCrumbs are self localizing and have several methods for initial location determination based on the environment they are placed in. GPS is not required for the BreadCrumbs to function and, once established, they can function as landmarks for autonomous robots by providing range data from radio signal strength with a path loss exponent determined through a Deep Deterministic Policy Gradient algorithm. The algorithm is designed such that a path loss exponent for each BreadCrumb location is learned during run time.

Genre

doctoral dissertations

Subjects--Topics

Electrical engineering

Degree

Ph.D.

Keywords

Deep Deterministic Policy Gradient
Localization
Path Loss Exponent
Received Signal Strength Indication
Trilateration

Subject Area

Electrical Engineering

Advisor(s)

Conrad, Jim

Committee Members

Chen, Chen
Smith-Orr, Courtney
Browne, Aidan

Degree Note

Thesis (Ph.D.)--University of North Carolina at Charlotte, 2021.

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

Grabowsky_uncc_0694D_12750

Permalink

http://hdl.handle.net/20.500.13093/etd:2606