| 基于强化学习的多机器人路径规划方法 |
| |
| 引用本文： | 宋一然.基于强化学习的多机器人路径规划方法[J].莆田学院学报,2006,13(2):38-41. |
| |
| 作者姓名： | 宋一然 |
| |
| 作者单位： | 湄洲湾职业技术学院,福建,莆田,351254 |
| |
| 基金项目： | 985工程建设项目(121503001) |
| |
| 摘 要： | 在未知环境下对多机器人路径规划问题进行了探讨,针对强化学习中再励函数的设计,提出了分段再励函数的设计方法,提高了机器人在未知环境下的学习效率。仿真结果表明了此方法的有效性。
|
| 关 键 词： | 多机器人 强化学习 再励函数 |
| 文章编号： | 1672-4143(2006)02-0038-04 |
| 收稿时间： | 2005-06-15 |
| 修稿时间： | 2005年6月15日 |
Path Planning of Multi-robots Systems Based on Reinforcement Learning |
| |
| SONG Yi-ran.Path Planning of Multi-robots Systems Based on Reinforcement Learning[J].journal of putian university,2006,13(2):38-41. |
| |
| Authors: | SONG Yi-ran |
| |
| Institution: | Meizhouwan Vocational Technology College, Putian 351254, China |
| |
| Abstract: | With the application of the multi-robots systems being trending to the unknown environment,we are going to research the reinforcement learning.In the paper,we bring forward the new structure of reinforcement function-reward function based subsection,which is designed by taking into account the information of sensor.The way is proven to improve the robot's learning efficiency by experiment. |
| |
| Keywords: | multi-robots reinforcement learning reinforcement function |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
