國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Multi-step motion planning: Applica...

Bretl, Timothy Wolfe.

Multi-step motion planning: Application to free-climbing robots.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Multi-step motion planning: Application to free-climbing robots.
作者:	Bretl, Timothy Wolfe.
面頁冊數:	117 p.
附註:	Adviser: Stephen M. Rock.
附註:	Source: Dissertation Abstracts International, Volume: 66-04, Section: B, page: 2170.
Contained By:	Dissertation Abstracts International66-04B.
標題:	Engineering, Aerospace.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3171640
ISBN:	9780542083693

Multi-step motion planning: Application to free-climbing robots.
Bretl, Timothy Wolfe.

Multi-step motion planning: Application to free-climbing robots. - 117 p.

Adviser: Stephen M. Rock.

Thesis (Ph.D.)--Stanford University, 2005.

A multi-step planning framework is presented for computing non-gaited, free-climbing motions. This framework derives from an analysis of a free-climbing robot's configuration space, which can be decomposed into constraint manifolds associated with each state of contact between the robot and its environment. An understanding of the adjacency between manifolds motivates a two-stage strategy that uses a candidate sequence of steps to direct the subsequent search for motions.

ISBN: 9780542083693Subjects--Topical Terms:

227946
Engineering, Aerospace.

Multi-step motion planning: Application to free-climbing robots.
LDR:03407nmm _2200301 _450 001 170654
005 20061228142210.5
008 090528s2005 eng d
020 $a 9780542083693
035 $a 00242684
040 $a UnM $c UnM
100 0 $a Bretl, Timothy Wolfe. $3 244682
245 1 0 $a Multi-step motion planning: Application to free-climbing robots.
300 $a 117 p.
500 $a Adviser: Stephen M. Rock.
500 $a Source: Dissertation Abstracts International, Volume: 66-04, Section: B, page: 2170.
502 $a Thesis (Ph.D.)--Stanford University, 2005.
520 # $a A multi-step planning framework is presented for computing non-gaited, free-climbing motions. This framework derives from an analysis of a free-climbing robot's configuration space, which can be decomposed into constraint manifolds associated with each state of contact between the robot and its environment. An understanding of the adjacency between manifolds motivates a two-stage strategy that uses a candidate sequence of steps to direct the subsequent search for motions.
520 # $a This dissertation addresses the problem of planning the motion of a multi-limbed robot to "free-climb" vertical rock surfaces. Free-climbing relies on natural features and friction (such as holes or protrusions) rather than special fixtures or tools. It requires strength, but more importantly it requires deliberate reasoning: not only must the robot decide how to adjust its posture to reach the next feature without falling, it must plan an entire sequence of steps, where each one might have future consequences. This process of reasoning is called multi-step planning.
520 # $a Three algorithms are developed to support the framework. The first algorithm reduces the amount of time required to plan each potential step, a large number of which must be considered over an entire multi-step search. It extends the probabilistic roadmap (PRM) approach based on an analysis of the interaction between balance and the topology of closed kinematic chains. The second algorithm addresses a problem with the PRM approach, that it is unable to distinguish challenging steps (which may be critical) from impossible ones. This algorithm detects impossible steps explicitly, using automated algebraic inference and machine learning. The third algorithm provides a fast constraint checker (on which the PRM approach depends), in particular a test of balance at the initially unknown number of sampled configurations associated with each step. It is a method of incremental precomputation, fast because it takes advantage of the sample distribution.
520 # $a Validation with real hardware was done with a four-limbed, free-climbing robot called LEMUR (developed by the Mechanical and Robotic Technologies Group at NASA-JPL). Using the multi-step planner presented in this dissertation, LEMUR free-climbed an indoor, near-vertical surface with artificial rock features. It did so with limited control and sensing, demonstrating that planning is absolutely critical.
590 $a School code: 0212.
650 # 0 $a Engineering, Aerospace. $3 227946
650 # 0 $a Computer Science. $3 212513
690 $a 0538
690 $a 0984
710 0 # $a Stanford University. $3 212607
773 0 # $g 66-04B. $t Dissertation Abstracts International
790 $a 0212
790 1 0 $a Rock, Stephen M., $e advisor
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://libsw.nuk.edu.tw:81/login?url=http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3171640 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3171640