東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

High-Speed Vision-Based Autonomous I...

Garcia, Adriano.

FindBook

Google Book

Amazon

博客來

High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation./
作者:	Garcia, Adriano.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	182 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-05, Section: B.
Contained By:	Dissertations Abstracts International82-05B.
標題:	Robotics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28030911
ISBN:	9798678160843

High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation.
Garcia, Adriano.

High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 182 p.

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

Thesis (Ph.D.)--State University of New York at Binghamton, 2020.

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

Autonomous vision-based navigation is one of the most rapidly expanding research fields in the industrial and academic worlds. A particularly new and potentially impactful research area involves the use of camera-enabled vision systems that permit quadrotor drones to navigate autonomously in structured indoor environments. Such autonomous navigation systems permit flights that are difficult to realize with a human controlling the drone based on visual cues in situations where the human controller cannot intervene in a timely or reliable fashion to deal with rapidly evolving flight situations. Autonomous techniques for navigating the drones in a reliable and consistent manner open up applications such as surveillance, situation assessments, mapping/exploration of unknown buildings and, in general, dealing with many emergencies. This dissertation develops a vision-based system technique for enabling the autonomous navigation of drones in building hallways based on the off-board processing of a monocular video stream transmitted from the drone's forward-facing camera.Indoor autonomous navigation using quadrotors is an especially challenging endeavor given the limitations and constraints due to the environment, the hardware, and the software currently available. As such, the work presented in this dissertation takes a holistic approach where the challenges and limitations of quadrotor aircraft are addressed via the use of proactive vision-based control algorithms that perceive the environment in order to understand it and develop a best course of action that permits uninterrupted autonomous flights at relatively high speeds. New methods to enable high-speed real-time flights are presented, where images acquired from a single front-facing camera onboard the quadrotor drone are used to extract environmental features that permit the drone to localize itself within the environment.The approach presented is purely vision-based where image processing, machine learning, and deep learning methods are leveraged to provide the external sensory mechanisms by which the environment is perceived and understood. The use of a single monocular front-facing camera affords several key benefits: no additional sensors or hardware are needed, no additional modification to the quadrotor platform is required, and precious onboard resources are saved due to reductions in payload and power consumption. In addition, since hardware requirements are low, our control software can easily be ported to other platforms.The autonomous system is validated through a wide variety of flight tests that demonstrate real-world consistent and reliable vision-based navigation in environments where few or no previous autonomous flights were feasible.

ISBN: 9798678160843Subjects--Topical Terms:

519753
Robotics.
Subjects--Index Terms:

Autonomous Navigation

High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation.
LDR:03991nmm a2200397 4500 001 2281716
005 20210920103349.5
008 220723s2020 ||||||||||||||||| ||eng d
020 $a 9798678160843
035 $a (MiAaPQ)AAI28030911
035 $a AAI28030911
040 $a MiAaPQ $c MiAaPQ
100 1 $a Garcia, Adriano. $3 3560425
245 1 0 $a High-Speed Vision-Based Autonomous Indoor Quadrotor Navigation.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 182 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-05, Section: B.
500 $a Advisor: Ghose, Kanad.
502 $a Thesis (Ph.D.)--State University of New York at Binghamton, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Autonomous vision-based navigation is one of the most rapidly expanding research fields in the industrial and academic worlds. A particularly new and potentially impactful research area involves the use of camera-enabled vision systems that permit quadrotor drones to navigate autonomously in structured indoor environments. Such autonomous navigation systems permit flights that are difficult to realize with a human controlling the drone based on visual cues in situations where the human controller cannot intervene in a timely or reliable fashion to deal with rapidly evolving flight situations. Autonomous techniques for navigating the drones in a reliable and consistent manner open up applications such as surveillance, situation assessments, mapping/exploration of unknown buildings and, in general, dealing with many emergencies. This dissertation develops a vision-based system technique for enabling the autonomous navigation of drones in building hallways based on the off-board processing of a monocular video stream transmitted from the drone's forward-facing camera.Indoor autonomous navigation using quadrotors is an especially challenging endeavor given the limitations and constraints due to the environment, the hardware, and the software currently available. As such, the work presented in this dissertation takes a holistic approach where the challenges and limitations of quadrotor aircraft are addressed via the use of proactive vision-based control algorithms that perceive the environment in order to understand it and develop a best course of action that permits uninterrupted autonomous flights at relatively high speeds. New methods to enable high-speed real-time flights are presented, where images acquired from a single front-facing camera onboard the quadrotor drone are used to extract environmental features that permit the drone to localize itself within the environment.The approach presented is purely vision-based where image processing, machine learning, and deep learning methods are leveraged to provide the external sensory mechanisms by which the environment is perceived and understood. The use of a single monocular front-facing camera affords several key benefits: no additional sensors or hardware are needed, no additional modification to the quadrotor platform is required, and precious onboard resources are saved due to reductions in payload and power consumption. In addition, since hardware requirements are low, our control software can easily be ported to other platforms.The autonomous system is validated through a wide variety of flight tests that demonstrate real-world consistent and reliable vision-based navigation in environments where few or no previous autonomous flights were feasible.
590 $a School code: 0792.
650 4 $a Robotics. $3 519753
650 4 $a Computer science. $3 523869
650 4 $a Automotive engineering. $3 2181195
650 4 $a Artificial intelligence. $3 516317
653 $a Autonomous Navigation
653 $a Autonomous Vehicles
653 $a Computer Vision
653 $a Convolutional Neural Networks
653 $a Machine Learning
653 $a Quadrotors
690 $a 0771
690 $a 0984
690 $a 0800
690 $a 0540
710 2 $a State University of New York at Binghamton. $b Computer Science. $3 1058053
773 0 $t Dissertations Abstracts International $g 82-05B.
790 $a 0792
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28030911