| 基于Q学习算法和BP神经网络的倒立摆控制 |
| |
| 引用本文： | 蒋国飞,吴沧浦.基于Q学习算法和BP神经网络的倒立摆控制[J].自动化学报,1998,24(5):662-666. |
| |
| 作者姓名： | 蒋国飞 吴沧浦 |
| |
| 作者单位： | 1.北京理工大学自控制系,北京 |
| |
| 摘 要： | Q学习是Watkins1]提出的求解信息不完全马尔可夫决策问题的一种强化学习方
法.将Q学习算法和BP神经网络有效结合,实现了状态未离散化的倒立摆的无模型学习控
制.仿真表明:该方法不仅能成功解决确定和随机倒立摆模型的平衡控制,而且和Anderson2]
的AHC(Adaptive Heuristic Critic)等方法相比,具有更好的学习效果.
|
| 关 键 词： | Q学习 BP网络 学习控制 倒立摆系统 高斯噪声 |
| 收稿时间： | 1997-1-22 |
| 修稿时间： | 1997-01-22 |
Learning to Control an Inverted Pendulum Using Q-Learning and Neural Networks |
| |
| JIANG GUOFEI,WU CANGPU.Learning to Control an Inverted Pendulum Using Q-Learning and Neural Networks[J].Acta Automatica Sinica,1998,24(5):662-666. |
| |
| Authors: | JIANG GUOFEI WU CANGPU |
| |
| Affiliation: | 1.Department of Automatic Control,Beijing Institute of Technology,Beijing |
| |
| Abstract: | Q-learning is a reinforcement learning method to solve Markovian decision problems
with incomplete information. This paper presents a novel method to control an inverted
pendulum with unquantized states by using Q-learning and neural networks. Simulation results
are included to show that the new method can not only balance the determined or
stochastic inverted pendulums successfully but also lead to a better effect of learning when
compared with Anderson's AHC method. |
| |
| Keywords: | Q-Learning BP neural network learning control inverted pendulum Gaussian noise |
| 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 |
