Event-triggered state estimator for stochastic systems with unknown inputs

This article studies the problem of state estimation for stochastic systems with unknown inputs. To reduce the communication cost from the sensor to the remote processor, an event-triggered communication mechanism is proposed in terms of an event generator function for the innovation vectors. The event-triggered estimator is developed by introducing an input term in the steady-state Kalman filter for the corresponding nominal system. The input gain matrix is determined by treating the nominal estimator error dynamics as the desired performance. It is shown that the estimation error is bounded in mean square under certain conditions. A numerical example is provided to verify the effectiveness of the proposed estimator.