Hybrid Vision-Based Navigation for Mobile Robots in Mixed Indoor/Outdoor Environments

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00223349" target="_blank" >RIV/68407700:21230/15:00223349 - isvavai.cz</a>

Result on the web

<a href="http://dx.doi.org/10.1016/j.patrec.2014.10.010" target="_blank" >http://dx.doi.org/10.1016/j.patrec.2014.10.010</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.patrec.2014.10.010" target="_blank" >10.1016/j.patrec.2014.10.010</a>

Result language

angličtina

Original language name

Hybrid Vision-Based Navigation for Mobile Robots in Mixed Indoor/Outdoor Environments

Original language description

In this paper we present a vision-based navigation system for mobile robots equipped with a single, off-the-shelf camera in mixed indoor/outdoor environments. A hybrid approach is proposed, based on the teach-and-replay technique, which combines a path-following and a feature-based navigation algorithm. We describe the navigation algorithms and show that both of them correct the robot's lateral displacement from the intended path. After that, we claim that even though neither of the methods explicitly estimates the robot position, the heading corrections themselves keep the robot position error bound. We show that combination of the methods outperforms the pure feature-based approach in terms of localization precision and that this combination reducesmap size and simplifies the learning phase. Experiments in mixed indoor/outdoor environments were carried out with a wheeled and a tracked mobile robots in order to demonstrate the validity and the benefits of the hybrid approach.

Czech name

—

Czech description

—

Type

J_x - Unclassified - Peer-reviewed scientific article (Jimp, Jsc and Jost)

CEP classification

JC - Computer hardware and software

OECD FORD branch

—

Project

<a href="/en/project/7AMB14AR015" target="_blank" >7AMB14AR015: Multi-robot autonomous systems</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2015

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

Pattern Recognition Letters

ISSN

0167-8655

e-ISSN

—

Volume of the periodical

53

Issue of the periodical within the volume

1

Country of publishing house

NL - THE KINGDOM OF THE NETHERLANDS

Number of pages

11

Pages from-to

118-128

UT code for WoS article

000349555500017

EID of the result in the Scopus database

2-s2.0-84922541097