BMI approach to design of networked control systems with decentralized event-triggering

In a networked control system (NCS) where plants and controllers are connected through communication networks, it is important to decrease the number of times of communications under preserving of control performances such as optimality and stability. From this viewpoint, event-triggered control has been studied so far. In event-triggered control, the measured signal is sent to the controller only when a triggering condition is satisfied. In this paper, the control problem with decentralized event-triggering conditions is investigated. Here, generalized event-triggering conditions with design parameters are utilized. First, the control problem studied here is formulated. Next, this original problem is reduced to the problem of minimizing an upper bound of the cost function in the original problem. Furthermore, the reduced problem is approximated by a BMI (bilinear matrix inequality) optimization problem. Finally, the proposed method is demonstrated by a numerical example.