Stochastic Analysis of NCSs with Random Delay and Data Dropout

Abstract

AbstractThe stability problem of closed-loop NCS in the presence of network delays and data packet drops has been addressed in [214, 211, 210, 149]. In [142, 196, 182], the problem of stochastic stability of network worked control systems with random time-delay or data dropout has been discussed, in which the random time-delay is modelled as a Markov process. Furthermore, in [213], the sensor-to-controller and controller-actuator delays are modeled as two Markov chains, and the resulting closed-loop systems are jump linear systems with two modes, where state feedback is proposed, the last received state is used for feedback again if there are delays longer than one sampling period or if there is a package loss. In the work of [117], the networked predictive control with the modified MPC is proposed, but the time-delay in the forward channel is constant. In this chapter, networked predictive control will be proposed for control of NCSs with random network delay in both the forward channel and feedback channel, where the network induced-delay is in the form of a Markov chain.This chapter is organized as follows. Section 14.2 gives the problem formulation and the scheme of predictive control and Section 14.3 presents the analysis of stochastic stability of the closed-loop system. Section 14.4 gives a numerical example and a practical experiment to demonstrate the effectiveness of the proposed method. Some conclusion remarks are given in Section 14.5.