非线性不确定系统的鲁棒拟人控制器设计  

Design of Robust Human-imitating Intelligent Controllers for Nonlinear Systems with Uncertainties

作　　者：廖道争[1] 王仁明[1] 孙文[2] 

机构地区：[1]三峡大学电气信息学院,湖北宜昌443002 [2]中国矿业大学(北京)资源学院,北京100083

出　　处：《三峡大学学报（自然科学版）》2006年第6期536-540,共5页Journal of China Three Gorges University：Natural Sciences

基　　金：航空基础科学基金资助项目(03E51002)

摘　　要：尽管拟人智能控制方法自提出以来取得了很多成功应用,但在系统存在不确定性的情况下,如何基于拟人智能控制方法设计系统的鲁棒控制器一直是个难题.针对仿射非线性不确定系统,结合二级倒立摆系统作为设计实例,提出了一种鲁棒拟人智能控制器设计方法.首先,基于被控对象的标称模型设计拟人控制律;然后在采用拟人控制律的基础上,应用积分滑模控制方案设计鲁棒补偿器,用以抑制模型不确定性对系统稳定的不利影响、提高系统的鲁棒性.最后,对二级倒立摆控制系统进行了仿真和实验验证,仿真和实验结果进一步证明了设计方案的有效性.Although the human-imitating intelligent control theory having several successful applications, the guarantee of the robustness of the designed human-imitating intelligent controllers when applied to uncertain systems is still a problem. A robust human-imitating controller for affine nonlinear uncertain systems was proposed with the double inverted pendulum as a design example. Firstly, a human-imitating control law was developed according to the nominal plant. Then, based on the human-imitating controller, a robust compensator was added to attenuate the influence of modeling errors and disturbances. The compensator was designed by means of integral sliding mode control theory. Applying the proposed control method to the double inverted pendulum, the results of simulations and experiments both verified the validity of this design.

关 键 词：拟人智能控制 积分滑模 鲁棒控制 非线性系统 二级倒立摆 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]