Bounding Uncertainty in State Estimation Under Dynamic Event-Triggered Communication

In the stochastic estimation context, the absence of measurement information at the state estimator during large intervals can cause a divergence in the uncertainty of the estimates. This issue is aggravated when strategies to reduce communication, such as event-triggering mechanisms (ETMs), are used if an appropriate design is not made. Particularly, dynamic ETMs (DETMs) may exhibit this problem, since they are designed to further reduce the number of communication instants. Motivated by this problem, we propose a novel state estimator that integrates discrete transmitted measurements and implicit information between events provided by the proposed DETM. Our proposal guarantees a uniformly bounded mean-squared error in the stochastic context, regardless of transmission instants. Moreover, compared to static ETMs, our proposal adaptively reduces the number of transmissions according to the behavior of the measured signal. Our proposal’s advantages are verified formally and through several numerical experiments.