采样与时延输出系统的状态观测研究

【作者】 李煜；

【导师】 王浩平；

【作者基本信息】 南京理工大学 ， 控制理论与控制工程， 2020， 硕士

【摘要】 近几年,随着网络通信与传感器技术的迅速发展,结合网络控制与视觉伺服的网络视觉伺服控制系统获得了越来越多的关注。与传统的点对点控制结构相比,网络视觉伺服控制系统有着诸多优势,例如连线简单,灵活度高,采样信息丰富等。然而,在实际应用中,其输出往往带有采样与时延,且系统受到非线性,未知输入扰动以及输出测量噪声的影响。本文综合考虑了上述采样、时延、未知输入扰动以及测量噪声等问题,将重点进行网络视觉伺服控制系统的状态观测器设计研究,以重构系统的连续无时延状态,具体的研究内容如下:(1)对网络视觉伺服控制系统的相关采样与时延特性进行分析,给出这类系统的线性模型。然后介绍本文的重要理论基础,即间断连续系统,给出这类系统的相关定义以及它的特性。(2)针对输出存在采样与时延的线性系统,考虑时延与采样间隔均为可变且未知的情况,设计了一种比例积分间断连续观测器,并对其进行了稳定性分析。这种观测器结合了间断连续系统,其优势在于,在每个采样时刻,观测器的状态都将重新初始化。这样可以防止估计误差累积,从而获得更准确的结果。最后将其与现有的间断连续观测器通过MATLAB/SIMULINK进行了数值仿真比较,证明了其有效性。(3)针对一类带有未知输入扰动且输出存在采样与时延的线性系统,考虑采样间隔固定但时延是未知且时变的情况,设计了一种扩张间断连续观测器。该观测器将时延看作网络扰动,对其进行估计并利用该估计值来补偿时延。文中给出了观测器的具体设计步骤并利用Lyapunov稳定性定理对观测器的稳定性进行了分析。接着将提出的观测器与Lyapunov-Krasovskii观测器进行了数值仿真比较,验证了观测器的有效性。(4)针对一类受到未知输入扰动和测量噪声的Lipschitz非线性系统,考虑采样间隔固定但时延是未知且时变的情况,设计了一种混杂扩张函数观测器。该观测器将输出时延和测量噪声等效看作系统的输出扰动,然后结合间断连续系统进行设计,同时给出了相应的稳定性证明。最后将其与连续观测器进行数值仿真比较,验证所提观测器的有效性。更多还原

【Abstract】 In recent years,with the rapid development of network communication and sensor technology,networked visual servo control systems combining network control and visual servo have gained more and more attention.This type of control system has many advantages,such as fewer wire connections,high flexibility,and more effective sampling information.However,in practical application,the output of the networked visual servo control system is often sampled and delayed,and the system is affected by non-linearity,unknown input disturbances and output measurement noise.This article comprehensively considers the issues of sampling,delay,unknown input disturbance,and measurement noise,and will focus on the state observer design of the networked visual servo control system to reconstruct the continuous state without delay of the system.The specific research content is as follows:(1)The relevant sampling and time delay characteristics of the networked vision servo control system are analyzed,and a linear model of this type of system is given.Then the important theoretical basis of this paper is introduced,namely piecewise continuous systems.The related definition and the characteristics of the system are given.(2)To deal with observer design problem for linear systems with sampled and delayed output,considering that the time delay and sampling interval are both variable and unknown,a proportional and integral piecewise continuous observer is designed and its stability is proved.This observer combines piecewise continuous systems with the advantage that the state of the observer will be re-initialized at each sampling moment,which can prevent estimation errors from accumulating and thus obtain more accurate results.Compared with the existing observer through numerical simulation by MATLAB/SIMULINK,the effectiveness of the proposed observer is proved.(3)For a class of linear systems with sampled and delayed output and subjected to unknown input disturbance,considering a situation where the sampling interval is fixed but the delay is unknown and time-varying,an extended piecewise continuous observer is designed.The observer treats the delay as the network disturbance,estimates it and uses it to compensate for the delay.The specific design steps of the observer are given and the stability of the observer is analyzed by using the Lyapunov stability theorem.Then,the proposed observer is compared with the Lyapunov-Krasovskii observer by numerical simulation,and the effectiveness of the observer is verified.(4)For a class of Lipschitz nonlinear system with sampled and delayed output,and is disturbed by unknown input and measurement noise,considering the case where the sampling interval is fixed but the delay is unknown and time-varying,a hybrid extended function observer is designed.The observer treats the output delay and measurement noise equivalent as the output disturbance of the system,and then designed with the concept of a piecewise continuous system.At the same time,the corresponding stability proof is given.Finally,it is compared with the continuous observer by numerical simulation to verify the effectiveness of the proposed observer.更多还原