Title :
Resource-oriented parallel planning
Author :
Lee, Sukhan ; Chung, Kyusik
Author_Institution :
Dept. of Electr. Eng.-Syst., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
A new paradigm for constructing a partially ordered plan network, referred to as resource-oriented parallel planning (RP2), is presented. In resource-oriented parallel planning, goal states are grouped into a collection of goal state subsets, each of which consists of those goal states associated with a particular resource. A subplan is generated for each goal state subset by planning the flow of the corresponding resource and thus satisfying the constraints specified by each of the goal state subsets. A complete plan is then constructed by synthesizing individual subplans based on the synchronization among subplans. A distinctive feature of RP2 is that each subplanner generates a conflict-free subplan by controlling the flow of a particular resource while synthesizing a complete plan in cooperation with other subplanners. The proposed scheme not only makes it possible to generate an optimal plan in terms of the maximum parallelism or the minimum depth of a plan network but also makes it easier to implement the concurrent generation of parallel plans in a parallel and distributed processing environment
Keywords :
algorithm theory; artificial intelligence; parallel algorithms; synchronisation; AI; conflict-free subplan; distributed processing; goal states; resource-oriented parallel planning; synchronization; Distributed processing; Frequency; Intelligent networks; Intelligent robots; Intelligent systems; Network synthesis; Parallel robots; Process planning; State-space methods;
Conference_Titel :
Tools for Artificial Intelligence, 1989. Architectures, Languages and Algorithms, IEEE International Workshop on
Conference_Location :
Fairfax, VA
Print_ISBN :
0-8186-1984-8
DOI :
10.1109/TAI.1989.65301