Stochastic model predictive control of multi-hop control networks with packet dropouts

A multi-hop control network (MHCN) is a control system in which plants and controllers are connected through a multi-hop wireless network. An MHCN is one of the typical examples in cyber-physical systems. In this paper, we propose a new method of stochastic model predictive control for MHCNs with packet dropouts. In the proposed method, a given MHCN is modeled by a stochastic switched system. Then, we consider the finite-time optimal control problem of minimizing the cost function given by the quadratic form with respect to the expected value of the state and the control input. This problem is reduced to a mixed integer programming problem. By solving it, both communication paths and control inputs are optimized simultaneously.