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

Adaptive neuro-fuzzy modeling and it...

University of Denver., Computer Engineering.

FindBook

Google Book

Amazon

博客來

Adaptive neuro-fuzzy modeling and its application to autonomous navigation.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Adaptive neuro-fuzzy modeling and its application to autonomous navigation./
作者:	Mitra, Arjun.
面頁冊數:	86 p.
附註:	Adviser: Irvin Jones.
Contained By:	Masters Abstracts International47-01.
標題:	Computer Science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1457000
ISBN:	9780549738374

Adaptive neuro-fuzzy modeling and its application to autonomous navigation.
Mitra, Arjun.

Adaptive neuro-fuzzy modeling and its application to autonomous navigation. - 86 p.

Adviser: Irvin Jones.

Thesis (M.S.)--University of Denver, 2008.

This thesis research project proposes modeling dynamic environments defined in Cartesian space utilizing the inherent linear separability property of neural networks. The activation function of the neurons fuzzifies the linear separation of points in Cartesian space. The composition of sets of linearly separated points create an entity or agent in the environment.

ISBN: 9780549738374Subjects--Topical Terms:

626642
Computer Science.

Adaptive neuro-fuzzy modeling and its application to autonomous navigation.
LDR:02572nmm 2200337 a 45 001 867183
005 20100802
008 100802s2008 ||||||||||||||||| ||eng d
020 $a 9780549738374
035 $a (UMI)AAI1457000
035 $a AAI1457000
040 $a UMI $c UMI
100 1 $a Mitra, Arjun. $3 1035895
245 1 0 $a Adaptive neuro-fuzzy modeling and its application to autonomous navigation.
300 $a 86 p.
500 $a Adviser: Irvin Jones.
500 $a Source: Masters Abstracts International, Volume: 47-01, page: 0414.
502 $a Thesis (M.S.)--University of Denver, 2008.
520 $a This thesis research project proposes modeling dynamic environments defined in Cartesian space utilizing the inherent linear separability property of neural networks. The activation function of the neurons fuzzifies the linear separation of points in Cartesian space. The composition of sets of linearly separated points create an entity or agent in the environment.
520 $a Due to the parallel nature of artificial neural networks this model is parallely decomposable and reduces the computation time either when incorporated on a hardware neural network platform or by utilizing multithreading in a simple processor. Hence, not only does this modeling methodology provide an efficient and viable mechanism to model the environment, but simultaneously creates a decision space (fuzzy membership), that provides meaningful information about the interaction of the entities in the universe of discourse.
520 $a To test this modeling methodology, it was applied to the domain of autonomous robot navigation in two dimensional Cartesian space. A number of computer simulations were carried out to test the feasibility of using the neuro-fuzzy modeling architecture with the Artificial Potential Fields (APF) method of robot navigation and the hybrid A-star (A*)---APF navigation strategy. The results of the simulations proved that the neuro-fuzzy modeling architecture can be successfully interfaced with the hybrid A-star (A*)---APF navigation strategy to provide a viable navigation heuristic for autonomous holonomic robot navigation in a dynamic environment.
590 $a School code: 0061.
650 4 $a Computer Science. $3 626642
650 4 $a Engineering, Robotics. $3 1018454
690 $a 0771
690 $a 0984
710 2 $a University of Denver. $b Computer Engineering. $3 1035894
773 0 $t Masters Abstracts International $g 47-01.
790 $a 0061
790 1 0 $a Jones, Irvin, $e advisor
790 1 0 $a Narapareddy, Vijaya L. $e committee member
790 1 0 $a Newman, Kim $e committee member
790 1 0 $a Salters, Roger $e committee member
791 $a M.S.
792 $a 2008
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1457000