摘要
气动人工肌肉的动态特性中存在着非常复杂的迟滞现象.目前对其迟滞特性的研究很不充分,甚至对其输入空间都难以确定.为此,建立了单自由度气动人工肌肉实验平台,利用分组数据处理神经网络独特的自组织特性,运用数据挖掘技术探索气动人工肌肉迟滞特性的输入空间.将自适应模糊小脑模型神经网络引入滑模控制,基于已确定的输入空间,在每个采样周期逼近迟滞力不断变化的动态值,在线实时补偿迟滞力的影响.实验结果验证了输入空间选取的合理性和有效性.
There exits hysteresis phenomenon in the dynamics of pneumatic artificial muscle（PAM）. By now the study on the hysteresis of PAM is so insufficient that even its input space is unknown. So, a sin gle degree of freedom PAM experiment facility was established. And the input space for the hysteresis of PAM was explored via a data mining technique group method of data handling（GMDH） neural network with the character of self organization. Based on the input space, an adaptive fuzzy cerebellar model articu- lation controller（CMAC） neural network was introduced into sliding mode control and the hysteresis of PAM was on-line compensated in real time. The experimental results suggest the rationality and effective- ness of the input space for the hysteresis of PAM.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2012年第6期931-935,共5页
Journal of Shanghai Jiaotong University
基金
上海市高校培养优秀青年教师科研专项基金(thc1005)
关键词
分组数据处理神经网络
气动人工肌肉
迟滞力
输入空间
group method of data handling （GMDH） neural network
pneumatic artificial muscle cerebel lar model articulation controller （PAM）
hysteresis
input space
