Control And Filtering Of Discrete Time Systems With Redundant Channels

Networked control system(NCS)has been widely developed and applied in many fields such as industrial automation system,unmanned driving field,remote surgery and so on.How-ever,due to influences of the network,it is inevitable that some networked induced phenomena occur,such as delay,packets loss,measurement attenuation and so on.These phenomena not only affect the dynamic performance of networked control systems,but also lead to instabili-ty.It can be significantly improved the probability of successfully delivering data packets by increasing the communication channels.Therefore,it is not only of theoretical significance but also of practical value to study the networked systems with redundant channels and design the corresponding controller/filter.In this thesis,based on Lyapunov stability theory and the method of stochastic analysis,the control and filtering problems of discrete-time systems with redundant channels are inves-tigatived.The main contents of this thesis are as follows:In the first part,a class of discrete-time systems with redundant channels are investigated.Firstly,a redundant channel is introduced to increase the probability of successfully deliver-ing measurements over the network,and two independent Bernoulli random variables are used to represent the packet dropouts phenomena in the main channel and redundant channel,re-spectively.Then,the stochastic stability of the system is proved by the method of Lyapunov function and the given H? performance index is satisfied.Finally,the design method of the observer-based controller is given by using the singular value decomposition technique.In the second part,the state estimation of Markov jump neural networks with random delay and redundant channels is considered.Firstly,the model of Markov jump neural networks with random delay and redundant channels is introduced.The random delay is bounded and subjects to known probability distribution.Then,by constructing a Lyapunov functional and in terms of linear matrix inequalities,the sufficient condition which guarantees the stochastic stability of the augmented estimated error system and the H? performance index is obtained.Finally,a model-dependent state estimator is designed.In the third part,it is extended to multiple redundant channels on the basis of the above first and two parts.Meanwhile,the dynamic output feedback control problem for nonlinear discrete time systems is considered.First of all,the probabilities of data packets loss in each comunication channel are expressed by a number of mutually independent Bernoulli random sequences.A nonlinear system model with multiple redundant channels is given.Then by constructing a suitable Lyapunov-like function,the finite-time stability and H? performance of the closed-loop system are analyzed.Finally,the method of designing dynamic output feedback controller for this kind of system is given.Finally,a summary and prospect of this thesis are given.