电动Stewart平台分散子系统不确定性加权函数选择  

Selection of uncertainty weighting function for decentralized subsystems of an electrical Stewart platform

作　　者：傅绍文[1] 姚郁[1] 韩蕾[2] 

机构地区：[1]哈尔滨工业大学控制与仿真中心,黑龙江哈尔滨150080 [2]中国民航大学航空自动化学院航空电子系,天津300300

出　　处：《控制理论与应用》2009年第4期415-419,共5页Control Theory & Applications

基　　金：国家自然科学基金资助项目(60674043)

摘　　要：以降低电动Stewart平台两输入两输出子系统不确定性描述的保守性为目的,提出一种不确定性加权函数选择方法.首先通过机理分析,确定以反馈形式来描述平台子系统中低频的结构不确定性,并得到具有3个待定参数的加权函数.然后给出标称模型输出与实验数据相匹配的条件,由此利用实验数据优化加权函数中的待定参数.最后为方便控制器设计,将系统不确定性描述化为统一的表达形式.所提方法在描述系统不确定性时,尽可能将模型中的已知部分分离出来,并通过实验数据优化加权函数中的待定参数,从而使得不确定性描述的保守性最小。To reduce the conservation in describing the uncertainty in the double input/double output （DIDO） subsystems of an electrical Stewart platform, we propose a method for selecting the weighting function. First, the structured uncertainty at low and middle frequencies is described in a feedback format based on the model analysis, and a weighting function of 3 undetermined parameters is obtained. Secondly, a criterion is proposed for matching the output with the experiment data; this criterion is then used for optimizing the parameters in the weighting function. Finally, the system uncertainty is transformed into a uniform expression for control design. Since the deterministic part of the system has been separated as far as possible, and the parameters of the weighting function have been optimized with using experiment data, the proposed method for uncertainty description has been rendered less conservative.

关 键 词：STEWART平台 模型不确定性 加权函数 鲁棒控制 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]