東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Linked to FindBook

Google Book

Amazon

博客來

Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints./
Author:	Woosley, Bradley.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
Description:	123 p.
Notes:	Source: Dissertations Abstracts International, Volume: 82-05, Section: B.
Contained By:	Dissertations Abstracts International82-05B.
Subject:	Robotics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28030889
ISBN:	9798684676192

Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints.
Woosley, Bradley.

Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 123 p.

Source: Dissertations Abstracts International, Volume: 82-05, Section: B.

Thesis (Ph.D.)--University of Nebraska at Omaha, 2020.

This item must not be sold to any third party vendors.

Multi-robot exploration of initially unknown environments is an important problem in many applications of robots including autonomous surveillance, unmanned search and rescue, extra-terrestrial exploration, and detection of hazardous phenomena such as chemical or toxic gas emissions. In this dissertation, we make four contributions towards making multi-robot exploration more adaptable to real-life settings: a) an intelligent exploration algorithm enables robots to integrate information and communication availability in the environment while selecting locations to explore, b) a novel algorithm that combines geometric approximation of regions explored by robots with a fast, tree-based, information caching technique to resolve conflicts between multiple robots that visit the same locations repeatedly, c) an approximation method for distance metric to improve the computation speed of selecting goal locations during the robots' exploration without significantly deteriorating the quality of collected information, and, d) a feasibility study into robots assuming another's mental state and track it down for rescue by a specialized robot. We have validated the proposed algorithms experimentally using both simulated and physical robots in different environment settings with different numbers of robots and different information distribution and communication availability profiles. Our results show that our proposed approaches can substantially improve the performance of multi-robot exploration under communication constraints: our information and communication integrated multi-robot exploration algorithm results in up to 25% reduction in distance to achieve a similar level of certainty of the mapped environment compared to only accounting for information gain; our region approximation and tree-based information caching algorithm for conflict resolution is able to resolve multi-robot conflicts with up to 62% fewer messages compared to a message flood approach; our utility approximation techniques provides an up to 7-fold improvement in the effectiveness of every second spent planning compared to our original method. Finally, our evaluation of assuming the robot's mental state shows promise for future exploration in developing robust algorithms to find missing robots.

ISBN: 9798684676192Subjects--Topical Terms:

519753
Robotics.
Subjects--Index Terms:

Alpha shapes

Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints.
LDR:03799nmm a2200457 4500 001 2344580
005 20220531064558.5
008 241004s2020 ||||||||||||||||| ||eng d
020 $a 9798684676192
035 $a (MiAaPQ)AAI28030889
035 $a AAI28030889
040 $a MiAaPQ $c MiAaPQ
100 1 $a Woosley, Bradley. $3 3683366
245 1 0 $a Multi-Robot Exploration for Improved Information Collection in Unknown Environments Under Communications Constraints.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 123 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-05, Section: B.
500 $a Advisor: Dasgupta, Prithviraj;Dorn, Brian.
502 $a Thesis (Ph.D.)--University of Nebraska at Omaha, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Multi-robot exploration of initially unknown environments is an important problem in many applications of robots including autonomous surveillance, unmanned search and rescue, extra-terrestrial exploration, and detection of hazardous phenomena such as chemical or toxic gas emissions. In this dissertation, we make four contributions towards making multi-robot exploration more adaptable to real-life settings: a) an intelligent exploration algorithm enables robots to integrate information and communication availability in the environment while selecting locations to explore, b) a novel algorithm that combines geometric approximation of regions explored by robots with a fast, tree-based, information caching technique to resolve conflicts between multiple robots that visit the same locations repeatedly, c) an approximation method for distance metric to improve the computation speed of selecting goal locations during the robots' exploration without significantly deteriorating the quality of collected information, and, d) a feasibility study into robots assuming another's mental state and track it down for rescue by a specialized robot. We have validated the proposed algorithms experimentally using both simulated and physical robots in different environment settings with different numbers of robots and different information distribution and communication availability profiles. Our results show that our proposed approaches can substantially improve the performance of multi-robot exploration under communication constraints: our information and communication integrated multi-robot exploration algorithm results in up to 25% reduction in distance to achieve a similar level of certainty of the mapped environment compared to only accounting for information gain; our region approximation and tree-based information caching algorithm for conflict resolution is able to resolve multi-robot conflicts with up to 62% fewer messages compared to a message flood approach; our utility approximation techniques provides an up to 7-fold improvement in the effectiveness of every second spent planning compared to our original method. Finally, our evaluation of assuming the robot's mental state shows promise for future exploration in developing robust algorithms to find missing robots.
590 $a School code: 1060.
650 4 $a Robotics. $3 519753
650 4 $a Artificial intelligence. $3 516317
650 4 $a Computer science. $3 523869
650 4 $a Technical communication. $3 3172863
650 4 $a Information technology. $3 532993
653 $a Alpha shapes
653 $a Communication constraints
653 $a Conflict resolution
653 $a Gaussian processes
653 $a Informed exploration
653 $a Multi-robot exploration
653 $a Autonomous surveillance
653 $a Extra-terrestrial exploration
653 $a Unmanned search and rescue
653 $a Chemical or toxic gas emissions
690 $a 0771
690 $a 0800
690 $a 0984
690 $a 0643
690 $a 0489
710 2 $a University of Nebraska at Omaha. $b Computer Science. $3 1020791
773 0 $t Dissertations Abstracts International $g 82-05B.
790 $a 1060
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28030889