摘要：AbstractWe consider the controller design for a linear time invariant system where the communication between the controller and the actuators is affected by correlated random packet losses. These packet losses are modelled as a finite length Markov chain. We present a method that takes advantage of the structure of the problem in order to design the optimal controller. The storage requirements for the optimal control laws however increase exponentially when longer packet loss histories have to be taken into account. We therefore present two sub-optimal methods that allow one to trade storage requirements for control performance. Finally, the performance in regard to storage and control optimality of the proposed designs are illustrated using simulations.
关键词：KeywordsOptimal controlWireless networksJump linear systemsHeuristicsStochastic SystemsNetworked SystemsDropouts
