基于EKF容错辨识的一阶倒立摆模糊控制  

Fuzzy Control of Inverted Pendulum System Based on Fault Tolerant Identification with EKF

作　　者：任玲[1] 张晓华[1] 

机构地区：[1]哈尔滨工业大学电气工程系,黑龙江哈尔滨150001

出　　处：《控制工程》2013年第S1期82-86,共5页Control Engineering of China

基　　金：国家自然科学基金资助项目(60875066)

摘　　要：针对一阶直线倒立摆运动控制系统给出了一种"单输入规则动态链接模糊控制算法(SIRMs)",算法简捷易于硬件实现,可使系统动态特性达到期望指标。针对一阶直线倒立摆系统小车质量产生较大突变时,仅依靠控制算法的鲁棒性已不能实现控制目标,系统动态性能变差甚至失控(容错能力不佳)的问题,采用扩展卡尔曼滤波方法 (EKF)对倒立摆系统中的小车质量进行参数辨识,根据辨识结果实时调整控制器参数,以使系统的动态性能达到期望指标,从而实现了系统的质量容错辨识与控制。仿真实验表明,结合EKF容错辨识算法的SIRMs控制器对于一阶直线倒立摆系统的运动控制具有良好的品质,并且对小车质量变化误差具有很好的容错能力。In this paper,a single input rules module dynamically linked fuzzy controller(SIRMs) is designed for an inverted pendulum system,and the fuzzy algorithm is simplified in order to make it easier to implement on the hardware platform.Changing the controller parameter improves dynamic characteristics of the system under the performance indexes.When the mass of inverted pendulum cart changes largely,the control purpose can not be realized only relying on robustness of the control algorithm.The dynamic performance of the system becomes poor and even the system loses stability,and the system is not tolerant for the car mass fault well.An extended Kalman filter(EKF) is used to estimate the fault-related parameter,while compensation of the fault is accomplished by using a switching controller with identified value.The simulation result illustrates that the inverted pendulum is controlled quickly and effectively with SIRMs control method combined with EKF,and this method has a good fault tolerant performance.

关 键 词：倒立摆 扩展卡尔曼滤波器(EKF) SIRMs模糊控制 容错辨识 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]