An approach to Action Planning Based on Simulated Annealing

被引:0
作者
Basilio Junior, Ricardo Rames [1 ]
Lopes, Carlos Roberto [1 ]
机构
[1] Univ Fed Uberlandia, Fac Comp, PPGCC, Uberlandia, MG, Brazil
来源
2012 XXXVIII CONFERENCIA LATINOAMERICANA EN INFORMATICA (CLEI) | 2012年
关键词
Fast Forward; Simulated Annealing; Enforced Hill-Climbing; ia planning; FF; SYSTEM;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
A number of new heuristic search methods have been developed in the field of automated planning over the last years. Enforced Hill-Climbing (EHC) is a method that has been frequently used in several AI planning systems. This method presents a better performance compared to other methods in many planning domains, but it has some weaknesses. In this paper, we investigate the use of Simulated Annealing in planners. In the experiments, this method shows significant results compared to a planner based on EHC.
引用
收藏
页数:7
相关论文
共 18 条
[1]   Fast forward planning by guided enforced hill climbing [J].
Akramifar, S. A. ;
Ghassem-Sani, G. .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2010, 23 (08) :1327-1339
[2]   Simulated annealing with restarts for the optimization of electromagnetic devices [J].
Alfonzetti, S ;
Dilettoso, E ;
Salerno, N .
IEEE TRANSACTIONS ON MAGNETICS, 2006, 42 (04) :1115-1118
[3]  
[Anonymous], 1998, Pddl the planning domain definition language
[4]   Macro-FF:: Improving AI planning with automatically learned macro-operators [J].
Botea, A ;
Enzenberger, M ;
Müller, M ;
Schaeffer, J .
JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2005, 24 :581-621
[5]  
Bryce D., 2006, ICAPS 2006 NOTES 5 I
[7]  
Coles A., 2008, C AI ED 23 AAAI C AR
[8]  
Coles A., 2007, J ARTIFICIAL INTELLI, P28
[9]   STRIPS - NEW APPROACH TO APPLICATION OF THEOREM PROVING TO PROBLEM SOLVING [J].
FIKES, RE ;
NILSSON, NJ .
ARTIFICIAL INTELLIGENCE, 1971, 2 (3-4) :189-208
[10]   COOLING SCHEDULES FOR OPTIMAL ANNEALING [J].
HAJEK, B .
MATHEMATICS OF OPERATIONS RESEARCH, 1988, 13 (02) :311-329