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

Real-time visual mosaicking and navi...

Stanford University.

Linked to FindBook

Google Book

Amazon

博客來

Real-time visual mosaicking and navigation on the seafloor.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Real-time visual mosaicking and navigation on the seafloor./
Author:	Richmond, Kristof.
Description:	110 p.
Notes:	Adviser: Stephen Rock.
Contained By:	Dissertation Abstracts International70-03B.
Subject:	Engineering, Aerospace. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3351469
ISBN:	9781109077339

Real-time visual mosaicking and navigation on the seafloor.
Richmond, Kristof.

Real-time visual mosaicking and navigation on the seafloor. - 110 p.

Adviser: Stephen Rock.

Thesis (Ph.D.)--Stanford University, 2009.

Remote robotic exploration holds vast potential for gaining knowledge about extreme environments accessible to humans only with great difficulty. Robotic explorers have been sent to other solar system bodies, and on this planet into inaccessible areas such as caves and volcanoes. In fact, the largest unexplored land area on earth lies hidden in the airless cold and intense pressure of the ocean depths. Exploration in the oceans is further hindered by water's high absorption of electromagnetic radiation, which both inhibits remote sensing from the surface, and limits communications with the bottom. The Earth's oceans thus provide an attractive target for developing remote exploration capabilities. As a result, numerous robotic vehicles now routinely survey this environment, from remotely operated vehicles piloted over tethers from the surface to torpedo-shaped autonomous underwater vehicles surveying the mid-waters.

ISBN: 9781109077339Subjects--Topical Terms:

1018395
Engineering, Aerospace.

Real-time visual mosaicking and navigation on the seafloor.
LDR:04863nmm 2200349 a 45 001 874338
005 20100824
008 100824s2009 ||||||||||||||||| ||eng d
020 $a 9781109077339
035 $a (UMI)AAI3351469
035 $a AAI3351469
040 $a UMI $c UMI
100 1 $a Richmond, Kristof. $3 1043604
245 1 0 $a Real-time visual mosaicking and navigation on the seafloor.
300 $a 110 p.
500 $a Adviser: Stephen Rock.
500 $a Source: Dissertation Abstracts International, Volume: 70-03, Section: B, page: 1922.
502 $a Thesis (Ph.D.)--Stanford University, 2009.
520 $a Remote robotic exploration holds vast potential for gaining knowledge about extreme environments accessible to humans only with great difficulty. Robotic explorers have been sent to other solar system bodies, and on this planet into inaccessible areas such as caves and volcanoes. In fact, the largest unexplored land area on earth lies hidden in the airless cold and intense pressure of the ocean depths. Exploration in the oceans is further hindered by water's high absorption of electromagnetic radiation, which both inhibits remote sensing from the surface, and limits communications with the bottom. The Earth's oceans thus provide an attractive target for developing remote exploration capabilities. As a result, numerous robotic vehicles now routinely survey this environment, from remotely operated vehicles piloted over tethers from the surface to torpedo-shaped autonomous underwater vehicles surveying the mid-waters.
520 $a However, these vehicles are limited in their ability to navigate relative to their environment. This limits their ability to return to sites with precision without the use of external navigation aids, and to maneuver near and interact with objects autonomously in the water and on the sea floor. The enabling of environment-relative positioning on fully autonomous underwater vehicles will greatly extend their power and utility for remote exploration in the furthest reaches of the Earth's waters---even under ice and under ground---and eventually in extraterrestrial liquid environments such as Europa's oceans.
520 $a This thesis presents an operational, fielded system for visual navigation of underwater robotic vehicles in unexplored areas of the seafloor. The system does not depend on external sensing systems, using only instruments on board the vehicle. As an area is explored, a camera is used to capture images and a composite view, or visual mosaic, of the ocean bottom is created in real time. Side-to-side visual registration of images is combined with dead-reckoned navigation information in a framework allowing the creation and updating of large, locally consistent mosaics. These mosaics are used as maps in which the vehicle can navigate and localize itself with respect to points in the environment.
520 $a The system achieves real-time performance in several ways. First, wherever possible, direct sensing of motion parameters is used in place of extracting them from visual data. Second, trajectories are chosen to enable a hierarchical search for side-to-side links which limits the amount of searching performed without sacrificing robustness. Finally, the map estimation is formulated as a sparse, linear information filter allowing rapid updating of large maps.
520 $a The visual navigation enabled by the work in this thesis represents a new capability for remotely operated vehicles, and an enabling capability for a new generation of autonomous vehicles which explore and interact with remote, unknown and unstructured underwater environments. The real-time mosaic can be used on current tethered vehicles to create pilot aids and provide a vehicle user with situational awareness of the local environment and the position of the vehicle within it. For autonomous vehicles, the visual navigation system enables precise environment-relative positioning and mapping, without requiring external navigation systems, opening the way for ever-expanding autonomous exploration capabilities.
520 $a The utility of this system was demonstrated in the field at sites of scientific interest using the ROVs Ventana and Tiburon operated by the Monterey Bay Aquarium Research Institute. A number of sites in and around Monterey Bay, California were mosaicked using the system, culminating in a complete imaging of the wreck site of the USS Macon , where real-time visual mosaics containing thousands of images were generated while navigating using only sensor systems on board the vehicle.
590 $a School code: 0212.
650 4 $a Engineering, Aerospace. $3 1018395
650 4 $a Engineering, Marine and Ocean. $3 1019064
650 4 $a Engineering, Robotics. $3 1018454
690 $a 0538
690 $a 0547
690 $a 0771
710 2 $a Stanford University. $3 754827
773 0 $t Dissertation Abstracts International $g 70-03B.
790 $a 0212
790 1 0 $a Rock, Stephen, $e advisor
791 $a Ph.D.
792 $a 2009
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3351469