Robust Filtering For Two Classes Of Network-based Uncertain Systems

State filtering has been attracting much attention as one of important problems in control theory and practical engineering applications. The filter actually can be explained as the states estimation of systems subject to disturbance based on past measurements, which significantly resolves the issues and phenomenon in communication, spaceflight, aviation, radar, biomedicine, image processing, target tracking and pattern recognition, etc. However, in many practical applications, since the statistical properties of exogenous noise signals are rarely known and there exist uncertainties in system model, the robust filtering problem attracts more and more interests. On the other hand, as the insertion of network in many industry systems, communication delay, data packet dropout and other problems occur inevitably in the control loops, and such factors often lower the performance of a system or even cause instability. Hence, the network-based filter design for uncertain systems is an important research issue which is with obvious theoretical significance and engineering value.The main results in the dissertation are as follows:In chapter 1, the basic concept of filtering and the detailed summary of the development process of filtering theory are introduced, respectively. Moreover, the significance of this thesis is presented not only in theory but also in applications. And the corresponding development at both abroad and home is recalled briefly.In chapter 2, the network-based filtering problem for uncertain systems is investigated. The networked system is modeled as a switched system with time-delay by the delay input approach, and such a system consists of two subsystems. Using the piecewise Lyapunov function approach, Jensen's integral inequality and the average dwell time method, a sufficient condition is obtained to guarantee that the filtering error system is exponentially stable, and the filter design method is presented in terms of linear matrix inequalities (LMIs). Finally, a numerical example is provided to illustrate the effectiveness of the proposed filter design method.In chapter 3, we are concerned with the design problem of robust filter for a class of discrete time systems with parameter uncertainties belonging to a polytope. In the network environment with network-induced delay and packed dropout, the networked filtering system is modeled as a discrete-time switched system. A LMI-based procedure is proposed for designing robust filter by Lyapunov stability theory and switched system theory, which guarantees the filtering error system to be exponentially stable. Finally, a numerical example is given to demonstrate the effectiveness of proposed method.In chapter 4, the main results of the thesis are concluded, and some research issues in future are proposed.