Tracking control of network distributed systems in presence of variable time delay and loss of information

Abstract

This paper deals with the control of network distributed systems which has been at the centre of interest in a wide area of research in the last few year. The control of such systems is very difficult because the communication networks inevitably introduce variable time delays and possible lost of samples. In particular, it is proposed an extension of the approach, derived in the contest of the optimal stochastic regulator problem [1], [2], to the remote tracking problem considering a distributed control system in which the signals from the transducers to the controller and from the controller to the actuator are transmitted through a communication network with variable delays and possible lost samples. The proposed approach uses timestamps, timeout and an one-step optimal predictor in the case of consecutive timeout. Simulation tests are carried out considering a first order system with the aim of validation of the proposed approach.