| 基于增强式学习的微小型移动机器人位姿控制 |
| |
| 引用本文： | 李小海,肖永利,程君实.基于增强式学习的微小型移动机器人位姿控制[J].上海交通大学学报,2000,34(11):1478-1482. |
| |
| 作者姓名： | 李小海 肖永利 程君实 |
| |
| 作者单位： | 上海交通大学,信息存储研究中心,上海,200030 |
| |
| 基金项目： | 国家自然科学基金重点资助项目(69889050) |
| |
| 摘 要： | 根据自行设计的两轮独立驱动的微小型移动机器人的特点，利用增强式的学习方法，对输入量进行分层模糊化以及确定模糊区域，然后选取相应区域中具有重大适应值的动作控制机器人移动，得到了该机器人在无约束空间中时间优化的移动路径，实现了该移动路径的自动获得，仿真和实验均得到了满意的结果。
|
| 关 键 词： | 移动机器人 位姿控制 增强式产 分层模糊化 |
| 文章编号： | 1006-2467(2000)11-1478-05 |
| 修稿时间： | 1999年11月12 |
Control of a Micro-Mobile Robot for Position and Orientation Problem Based on Reinforcement Learning |
| |
| LI Xiao-hai,XIAO Yong-li,CHENG Jun-shi.Control of a Micro-Mobile Robot for Position and Orientation Problem Based on Reinforcement Learning[J].Journal of Shanghai Jiaotong University,2000,34(11):1478-1482. |
| |
| Authors: | LI Xiao-hai XIAO Yong-li CHENG Jun-shi |
| |
| Abstract: | According to the characteristic of the micro mobile robot with two independent wheels, this paper proposed a control method for its position and orientation problem. This method is based on reinforcement learning. It uses the hierarchical fuzziness of the inputs to get the fuzzy area. Every fuzzy area has three kinds of actions. The micro mobile robot moves according to the action with the maximal performance in this fuzzy area. This method gets the moving route automatically within the optimal time in unrestraint space. The imitation and experimental results show this method is feasible. |
| |
| Keywords: | mobile robot position and orientation controlling reinforcement learning hierarchical fuzzy |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
