| 基于行为建模的机器人混合模糊-神经控制方法 |
| |
| 引用本文： | 谭群华,李伟.基于行为建模的机器人混合模糊-神经控制方法[J].清华大学学报(自然科学版),1997(3). |
| |
| 作者姓名： | 谭群华 李伟 |
| |
| 作者单位： | 清华大学计算机科学与技术系(谭群华)，智能技术与系统国家重点实验室(李伟) |
| |
| 摘 要： | 本文提出了一种设计机器人模糊－神经控制器的新方法。因为模糊逻辑控制器的控制信号是由系统的响应行为而不是由其分析模型决定的，所以机器人的开环响应可以用一系列二阶系统来描述。然后，用Ｎｅｌｄｅｒ－Ｍｅａｄ单纯形算法离线优化与该系统相关的模糊逻辑控制器的参数，并用人工神经网络来训练开环响应与这些优化了的参数之间的匹配关系。该方法的优点在于当它用于设计模糊－神经控制器时，在自适应过程中不必考虑收敛问题
|
| 关 键 词： | 行为建模 模糊控制 非线性控制 机械手 |
Behavior based neuro fuzzy control method for robot |
| |
| Tan Qunhua,Li Wei.Behavior based neuro fuzzy control method for robot[J].Journal of Tsinghua University(Science and Technology),1997(3). |
| |
| Authors: | Tan Qunhua Li Wei |
| |
| Institution: | Tan Qunhua,Li Wei Department of Computer Science and Technology,Tsinghua University, State Key Laboratory of Intelligent Technology and Systems,Beijing 100084 |
| |
| Abstract: | This paper presents a new approach to design a neuro fuzzy controller for robot. Because the signal processed by fuzzy controller is based on system behavior rather than its analytical model, open loop responses of a robot are described by a set of second order systems. Then, the parameters of fuzzy logic controller is off line optimized by Nealder Mead's Simplex Algorithm, and use neuro network to train mapping relationship between open responses and parameters. The key advantage of this method is that when it is used to design Neuro Fuzzy controller , the convergence problem is not in necessity. |
| |
| Keywords: | behavior based modeling fuzzy control nonlinear control manipulator robot |
| 本文献已被 CNKI 等数据库收录！ |
