Title :
Fast planning algorithm based on stack of action sets and high efficiency planning graph
Author :
Lu, Ying-Hua ; Hou, Gang ; Gu, Wen-xiang
Author_Institution :
Sch. of Comput., North East Normal Univ., Changchun, China
Abstract :
Based on the study and analysis of the traditional planning graph and its algorithm, this paper presents a novel planning graph, named high efficiency planning graph. It is a directed, leveled graph with three kinds of nodes and four kinds of edges which can generate the valid plan directly at the same time with graph extending, without backward-chaining searching for the valid plan. Meanwhile, the mutex relationships between actions are redefined in this paper. Reference graph and stack of action sets are used to observably improve the speed of building the valid plan.
Keywords :
directed graphs; planning (artificial intelligence); search problems; backward-chaining searching; fast planning algorithm; high efficiency planning graph; mutex relationships; reference graph; stack of action sets; Algorithm design and analysis; Artificial intelligence; Cybernetics; Educational institutions; History; Intelligent systems; Machine learning; Efficiency Planning Graph; Graphplan; Planning; Reference Graph; Stack of Action Sets;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527167