计算机科学 ›› 2015, Vol. 42 ›› Issue (4): 217-220.doi: 10.11896/j.issn.1002-137X.2015.04.044
李 洋,文中华,伍小辉,劳佳琪
LI Yang, WEN Zhong-hua, WU Xiao-hui and LAO Jia-qi
摘要： 现实世界中,动作的执行通常都要耗费一定的代价,且由于外界环境的干扰,动作执行后的结果具有不确定性。针对这一问题,对不确定状态转移系统的动作赋予权值,使用概率分布表示状态转换的随机性,提出了强循环规划解的期望权值,并且设计了求最小期望权值强循环规划解的方法。该方法的主要思想是使用深度优先搜索求出规划问题的所有强循环规划解,再将强循环规划解分别转换成 以 状态到目标状态的期望权值为变元的线性方程组,最后使用高斯消元法解方程组,从而找出最小期望权值强循环规划解。
| [1] Kuter U,Nau D,Reisner,et al.Using Classical Planners to Solve Nondeterministic Planning Problems[C]∥Proc.of the 18th Int Conf on Automated Planning and Scheduling.Menlo Park,CA:AAAI press,2008 [2] Bertoli P,Cimatti A,Roveri M,et al.Strong planning under partial observability[J].Artificial Intelligence,2006,0(4/5):337-384 [3] Cimatti A,Roveri M,Traverso P.Automatic OBDD-based gene-ration of universal plans in nondeterministic domains[C]∥Proc.of the 15th National Conf on Artificial Intelligence.Menlo Park,CA:AAAI Press,1998:875-881 [4] Cimatti A,Pistore M,Roveri M,et al.Weak,strong,and strong cyclic planning via symbolic model checking[J].Artificial Intelligence,2003,147(1/2):35-84 [5] 文中华,黄巍,刘任任,等.模型检测规划中的状态分层方法[J].软件学报,2009,20(4):858-869 [6] Oh J,Meneguzzi F,Sycara K.Probabilistic plan recognition for intelligent information agents[C]∥Proc.ICAART.2011 [7] Martín Y E,Moreno M D R,Smith D E.Probabilistic PlanGraph Heuristic for Probabilistic Planning[C]∥Proc.of the Twenty-Fifth AAAI Conference on Artificial Intelligence.2011:1774-1745 [8] Fu J,Ng V,Bastani F B,et al.Simple and fast strong cyclic planning for fully-observable nondeterministic planning problems[C]∥Proc.of the 22nd Int Joint Conf on Artificial Intelligence.Menlo Park,CA:AAAI Press,2011:1949-1954 [9] 周俊萍,殷明浩,谷文祥,等.部分可观察强规划中约减观察变量的研究[J].软件学报,2009,0(2):290-304 [10] 唐杰,文中华,汪泉,等.不确定可逆规划的强循环规划解[J].计算机研究与发展,2013,0(9):1970-1980 [11] 刘铭,徐杨,陈峥,等.基于Multi-agent系统的多飞行器协同路径规划方法的研究[J].计算机科学,2012,9(1):219-233 [12] 陈建林,文中华,马丽丽,等.一种求解最小权值强规划的方法[J].计算机工程,2011,7(17):167-171 |
| No related articles found! |
|
||
首页