| 移动机器人的自学习导航策略 |
| |
| 引用本文： | 李强,林良明,颜国正.移动机器人的自学习导航策略[J].上海交通大学学报,2001,35(5):779-782,786. |
| |
| 作者姓名： | 李强 林良明 颜国正 |
| |
| 作者单位： | 上海交通大学电子信息学院, |
| |
| 摘 要： | 针对标准Q学习收敛较慢的缺点，采用多步Q学习算法，为解决连续过程的学习问题，采用CMAC神经网络对连续状态空间进行泛化，讨论了基于CMAC的多步Q学习算法应用于导航系统的实现方法，并与其他方法进行仿真比较，结果了表明了该算法的有效性。
|
| 关 键 词： | 导航 Q学习 CMAC神经网络 移动机器人 |
| 文章编号： | 1006-2467(2001)05-0779-04 |
Self-Learning Navigation Strategy of Mobile Robot |
| |
| LI Qiang,LIN Liang ming,YAN Guo zheng.Self-Learning Navigation Strategy of Mobile Robot[J].Journal of Shanghai Jiaotong University,2001,35(5):779-782,786. |
| |
| Authors: | LI Qiang LIN Liang ming YAN Guo zheng |
| |
| Abstract: | The self learning navigation strategy of mobile robot in unknown environment based on Q learning was studied. A multi step Q learning algorithm was employed to overcome the slow convergence rate of standard Q learning, and CMAC neural network was used to generalize the continuous state space. The application of the algorithm to navigation system was given as well. The simulation and comparison with other algorithm were also given. The results demonstrate the efficiency of the algorithm. |
| |
| Keywords: | navigation Q learning CMAC neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
