stateobserver-控制科学与工程学院-浙江大学.pdf

Principle of Automatic Control State Space: Analysis and Design 浙江大学控制科学与工程学院 自动控制原理浙江大学控制学院2016 Outline of Section 6 ü State Observer for SISO System ü基本思想 ü全维观测器 ü分离原理 ü降维观测器 注：这部分内容在教材的第四版中没有相对应的章节， 但在第五版中已经增加，请参见：Ed.5 ：13.13, 13.14。 自动控制原理浙江大学控制学院2016 1. Basic concept •在第4 节讨论通过状态反馈u=Kx 进行极点任意配置时，总是假设所 有的状态都是可以获取的。 u State Eq. x Output Eq. y y = Cx x = Ax + Bu controller Conditions: (1) state variables must State feedback: u(t)=kx(t) be controllable; (2) state variables should be accessible. 自动控制原理浙江大学控制学院2016 1. Basic concept If state variables cannot be accessible, but the system is observable, the real-time state variables can be estimated by output and input of the system , as following Fig.. The estimator is called state observer. u 状态方程 x 输出方程 y y = Cx x = Ax + Bu ˆ x 控制器 状态观测器 State feedback: u(t)=kx(t) State observer l 确定性条件下的状态重构问题 è Luenberger观测器理论 l 噪声环节下的状态估计涉及随机最优估计理论 è Kalman滤波器理论 自动控制原理浙江大学控制学院2016 1. Basic concept • In many practical control systems it is physically or economically impractical to install all the transducers which would be necessary to measure all the states.Why? • The