| 连续时间MCP在紧致行动集上的最优策略 |
| |
| 引用本文： | 奚宏生,唐昊,殷保群.连续时间MCP在紧致行动集上的最优策略[J].自动化学报,2003,29(2):206-211. |
| |
| 作者姓名： | 奚宏生 唐昊 殷保群 |
| |
| 作者单位： | 1.中国科学技术大学自动化系,合肥 |
| |
| 基金项目： | NationalNaturalScienceFoundationofP .R .China (6 9974 0 37)，NationalHighPerformanceComput ingFoundationofP .R .China(0 0 2 0 8) |
| |
| 摘 要： | 文中研究了一类连续时间Markov控制过程(CTMCP)无穷水平平均代价性能的最优控
制决策问题.文章采用无穷小生成元和性能势的基本性质,直接导出了平均代价模型在紧致行动
集上的最优性方程及其解的存在性定理,提出了求解ε-最优平稳控制策略的数值迭代算法,并给
出了这种算法的收敛性证明.最后通过分析一个数值例子来说明这种方法的应用.
|
| 关 键 词： | 性能势 平均代价准则 紧致行动集 数值迭代 |
| 收稿时间： | 2001-10-8 |
Optimal Policies for a Continuous Time MCP with Compact Action Set |
| |
| XI Hong-sheng,TANG Hao,YIN Bao-qun.Optimal Policies for a Continuous Time MCP with Compact Action Set[J].Acta Automatica Sinica,2003,29(2):206-211. |
| |
| Authors: | XI Hong-sheng TANG Hao YIN Bao-qun |
| |
| Affiliation: | 1.Department of Automation,University of Science and Technology of China,Hefei |
| |
| Abstract: | In this paper, we study optimal policies for a class of continuous-time Markov control processes (CTMCPs) with infinite horizon average-cost criteria. Using the basic properties of infinitesimal generators and performance potentials, we give directly the optimality equation and establish the existence of solutions to this equation for the average-cost model on a compact action set. A fast value iteration algorithm, which leads to an ε-optimal stationary policy, is proposed and the convergence of this algorithm is studied. Finally, we provide one numerical example to show applications of the proposed method. |
| |
| Keywords: | Performance potentials average cost criteria compact action set value iteration |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
