C. Nielsen and L. E. Kavraki, "A two-level fuzzy prm for manipulation planning," in Proceedings of The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE Press (Refereed), November 2000, pp. 1716--1722.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....of the manipulation task into such elementary motion planning sub tasks, and the solution of each one of them. Most of existing manipulation planning algorithms assume that finite sets of stable placement and possible grasps of a movable object are given in the definition of the problem (e.g. [1, 2, 4, 8, 11]) Consequently, a part of the manipulation task decomposition problem is thus resolved by the user. Returning to the example, getting the bar out of the cage requires a large number of precise placements and grasps that must be input data for these algorithms. Dealing with con tinuous grasp and ....

Ch. Nielsen and L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE Int. Conference on Intelligent Robots and systems, 2000.

....In PRM planning, a signi cant amount of time is wasted in showing that candidate segments between sampled milestones are actually colliding. In [8, 22] it was shown that in practice the prior probability of a segment to be colliding increases with its length in c space. The planners in [20, 22] exploit this result to test multi segment paths, by maintaining a priority queue of (sub )segments sorted by decreasing length. Here, we can take advantage that we also know bounds on workspace distances and on lengths of traced curves for each (sub )segment. Intuitively,two bodies are more ....

Ch. Nielsen and L. E. Kavraki. A two-level fuzzy PRM for manipulation planning. In ##### ## ### ######## ############# ########## ## ########### ###### ### ####### ######, Japan, 2000.

....ferm ees [12, 27] corps d eformables [3] ffl des contraintes sur les mouvements: syst emes non holonomes [26] contraintes dynamiques [28, 19] ffl de probl emes n ecessitant l exploration d espaces hautement dimensionn es: planification multirobots [38] ou de taches de manipulation. [1, 31]. Dans cet article nous pr esentons une synth ese de travaux sur les m ethodes PRM (Probabilistic Roadmap Methods) qui trouvent leur origine a Stanford et Utrecht et qui ont par la suite donn e lieu a de nombreuses variantes ainsi qu a la plupart des extensions mentionn ees ci dessus. La ....

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE International Conference on Intelligent Robots and systems, Japan, 2000.

....planning [2] see also Chapter 11 of Latombe s book [13] shows that a solution is an alternate sequence of elementary motions, called transit and transfer paths, and separated by grasp and ungrasp operations. Several algorithms have been proposed to solve different instances of problems [2, 9, 3, 5, 1, 17]. While the existing planners generally assume of discrete number grasps and of stable placements of the movable objects, dealing with continuous sets may allow more sophisticated planners to be designed. Hence, discrete sets suppose that the knowledge initial provided by the user contains some of ....

.... to deal in realistic situations with redundant manipulators for which each grasp possibly corresponds to an infinite number of robot configurations (i.e. the cardinality of CP CG is infinite) More recently, another practical manipulation planner using the PRM framework [9, 18] was proposed in [17]. The algorithm assumes discrete grasps and placements of a single movable object. As in [2] a manipulation graph between discrete configurations of CP CG is constructed but connections are computed using a Fuzzy PRM planner that builds a roadmap with edges annotated by a probability of ....

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE International Conference on Intelligent Robots and systems, Japan, 2000.

....in the regions determined promising by the initial coarse processing, which can save a signi cant amount of processing time. A similar philosophy 2 (a) b) Figure 1: a) control roadmap, and (b) the resulting approximate roadmap for an environment. is proposed in the Lazy PRM [6] and Fuzzy PRM [16] methods, which advocate an even coarser validation than c prm either no validation at all [6] i.e. accepting all roadmap nodes and edges in the construction phase) or validating nodes but not edges [16] Our experience is that some, even very limited, validation of both nodes and edges is ....

....approximate roadmap for an environment. is proposed in the Lazy PRM [6] and Fuzzy PRM [16] methods, which advocate an even coarser validation than c prm either no validation at all [6] i.e. accepting all roadmap nodes and edges in the construction phase) or validating nodes but not edges [16]. Our experience is that some, even very limited, validation of both nodes and edges is essential to provide an estimate of the free space to guide path selection, which in turn can signi cantly reduce query costs. In the following, we describe the details of the roadmap construction and the ....

C. L. Nielsen and L. E. Kavraki. A two level fuzzy prm for manipulation planning. Technical Report TR2000-365, Computer Science, Rice University, Houston, TX, 2000.

....that also support the choice and the use of several handling devices for carrying out objects within an industrial installation. This problem referred to as the manipulation planning problem [2] remains a practical challenge because of its additional complexity, although several promising results [21, 1, 27] have been recently obtained using probabilistic techniques. Acknowledgments: The practical results have been obtained from the motion planning software Move3D developed in collaboration with C. Nissoux. We also thank EDF and Cadcentre for providing models of industrial scenes. ....

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE Int. Conference on Intelligent Robots and systems, Japan, 2000.

.... For example, problems where the solution path must traverse narrow passages in C space [12] As a result, a number of prm variants speci cally targeted at this problem have been proposed, e.g. 1, 6, 12] Also, some novel approaches for improving prm eciency have shown promising results [5, 20] (these methods do not address the narrow passage problem) Unfortunately, however, prm solutions to many problems still take prohibitively long. Another shortcoming of prms is that while they are very good at nding a path, they do not support applications which might impose particular, ....

....O(nkd) validity checks the majority of which are not needed for any particular solution path. Our strategy of postponing validation leads to faster roadmap construction, while incurring only slightly higher query times. As discussed below, similar strategies have been used by other researchers [5, 20]. 1 An important bene t of our approach is that it gives one the ability to customize the roadmap in accordance with any speci ed, variable, query requirements. For example, it might be speci ed that the solution path should maintain a certain clearance from the obstacles, that the robot s ....

[Article contains additional citation context not shown here]

C. L. Nielsen and L. E. Kavraki. A two level fuzzy prm for manipulation planning. Technical Report TR2000-365, Computer Science, Rice University, Houston, TX, 2000.

....that also support the choice and the use of several handling devices for carrying out objects within an industrial installation. This problem referred to as the manipulation planning problem [2] remains a practical challenge because of its additional complexity, although several promising results [22, 1, 28] have been recently obtained using probabilistic techniques. Acknowledgment The practical results have been obtained from the generic motion planning software platform Move3D developed in collaboration with C. Nissoux. We also thank EDF and Cadcentre for providing models of industrial scenes. ....

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE International Conference on Intelligent Robots and systems, Japan, 2000.

.... has been the topic of extensive research over the last decade [1, 9, 20, 10] The complexity of the problem is high and several versions of it have been shown PSPACE hard [20] Interesting applications and extensions of the problem exist in planning for robots that can modify their environments [11, 23] and flexible robots [3] planning for graphics and simulation [18] planning for virtual prototyping [7] and planning for medical [25] and pharmaceutical [8] applications. This paper concentrates on the analysis of PRM [17, 14] Since 1994, when PRM was invented, several researchers have ....

.... robots with many degrees of freedom, several variations of the method have been developed (e.g. 2] several planners that have been influenced by PRM have been introduced (e.g. 13, 19] and several extensions of the basic path planning problem have been solved with PRM based methods (e.g. [23]) The experimental success of the planner has motivated many researchers to seek a theoretical basis for explaining its performance and relative successes in this direction have been reported, among others, in [5, 6, 16, 15, 28, 13, 26] This paper presents a further extension in this direction ....

C. Nielsen and L. Kavraki. A two level fuzzy PRM for manipulation planning. In IEEE/RSJ Int. Workshop on Intelligent Robots & Systems (IROS), Japan, 2000.

.... has been the topic of extensive research over the last decade [1, 8, 16] The complexity of the prob lem is high and several versions of it have been shown PSPACE hard [16] Interesting applications and ex tensions of the problem exist in planning for robots that can modify their environments [9, 18] and flexible robots [3] planning for graphics and simulation [15] planning for virtual prototyping [6] and planning for medical [20] and pharmaceutical [7] applications. This paper concentrates on the analysis of PRM [14, 11] Since 1994, when PRM was invented, several researchers have ....

.... for robots with many degrees of freedom, several variations of the method have been developed (e.g. 2] several planners that bare resemblances with PRM have been introduced (e.g. 10] and sev eral extensions of the basic path planning problem have been solved with PRM based methods (e.g. [18]) The experimental success of the planner has motivated many researchers to seek a theoretical basis for explaining its performance and relative successes in this direction have been reported, among others, in [5, 13, 12, 23, 10, 21] This paper presents a further extension in this direction by ....

C. Nielsen and L. Kavraki. A two level fuzzy PRM for manipulation planning. In IEEEJRSJ Int. Workshop on Intelligent Robots 4 Systems (IROS), Japan, 2000.

....PRMs, we do not build a roadmap of feasible paths, but rather a roadmap of paths assumed to be feasible. The idea is to lazily evaluate the feasibility of the roadmap as planning queries are processed. Similar ideas about lazy evaluation have been developed con currently but independently in [25] in a planner called Fuzzy PRM. In other words, let qinit, q9oal, and a number of uni formly distributed points form nodes in a roadmap, and connect by edges each pair of nodes being sufficiently close together. Given a procedure that estimates the length of a path, we find a shortest feasible ....

C.L. Nielsen and L.E. Kavraki. A two level fuzzy PRM for manipulation planning. Technical Report TR2000.

....paper. Section 4 gives a description of the basic PRM, a number of variations of the algorithm, and other closely related algorithms. In Section 5 we describe Lazy PRM. We draw our discussion from [6, 7, 8] Related ideas about lazy evaluation have been developed concurrently and independently in [35]. An analysis of the planner is given in Section 6 while experimental results are given in Section 7. We conclude in Section 8 with a discussion of the capabilities and limitations of Lazy PRM. Some of our comments apply to randomized approaches to path planning in general. COMPLEXITY ISSUES In ....

....to at least two components of the roadmap, or if it does not see any other node. In the former case the components are merged, and in the latter case a new component is created. Variations of PRM have also been used for manipulation planning and for robots with closed kinematic chains, see [16, 31, 35]. 5. DESIGNING AN EFFICIENT RANDOMIZED PLANNER USING A LAZY EVALUATION SCHEME 5.1 MOTIVATION In many applications, the configuration space changes frequently. For example, as soon as the robot changes tools, grasps or deforms an object, or when a new obstacle enters the workspace, the ....

C.L. Nielsen and L.E. Kavraki. A two level fuzzy PRM for manip- ulation planning. Technical Report TR2000-365, Rice University, 2000.

....These cases are handled to a certain extent, but neither these cases nor the narrow passage problem (see [2, 8, 18] are our main objectives. This paper extends the results presented in [7] and [6] Related ideas about lazy evaluation has been developed concurrently and independently in [37]. Lazy PRM is described in detail in Section 3. Its performance is theoretically analyzed in Section 4, and experimentally evaluated in Section 5 using a real industrial environment. 2 Path Planning Techniques Path planning is becoming increasingly important in automated manufacturing industry ....

....to at least two components of the roadmap, or if it does not see any other node. In the former case the components are merged, and in the latter case a new component is created. Variations of PRM have also been used for manipulation planning and for robots with closed kinematic chains, see [16, 33, 37]. The general theme for roadmap algorithms is to construct a network of paths veri ed to be collision free by a local planner. Unfortunately, it is dicult to nd a global strategy that can use these local planners eciently in order to avoid traps and dead ends. In industrial environments, with ....

C.L. Nielsen and L.E. Kavraki. A two level fuzzy PRM for manipulation planning. Technical Report TR2000-365, Rice University, 2000.

....PRMs, we do not build a roadmap of feasible paths, but rather a roadmap of paths assumed to be feasible. The idea is to lazily evaluate the feasibility of the roadmap as planning queries are processed. Similar ideas about lazy evaluation have been developed concurrently but independently in [25] in a planner called Fuzzy PRM. In other words, let q init , q goal , and a number of uniformly distributed points form nodes in a roadmap, and connect by edges each pair of nodes being suciently close together. Given a procedure that estimates the length of a path, we nd a shortest feasible ....

C.L. Nielsen and L.E. Kavraki. A two level fuzzy PRM for manipulation planning. Technical Report TR2000365, Rice University, 2000.

No context found.

C. Nielsen and L. E. Kavraki, "A two-level fuzzy prm for manipulation planning," in Proceedings of The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE Press (Refereed), November 2000, pp. 1716--1722.

No context found.

C. Nielsen and L. Kavraki. A Two-Level Fuzzy PRM for Manipulation Planning. Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, 2000

No context found.

C. L. Nielsen and L. E. Kavraki. A two level fuzzy prm for manipulation planning. Technical Report TR2000-365, Computer Science, Rice University, Houston, TX, 2000.

No context found.

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE Int. Conference on Intelligent Robots and systems, Japan, 2000.

No context found.

Ch. Nielsen, L. Kavraki. A two-level fuzzy PRM for manipulation planning. In IEEE Int. Conf. on Intelligent Robots and Systems, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC