Observability Gramian for Bayesian Inference in Nonlinear Systems With Its Industrial Application

In this letter, we present a novel (empirical) observability Gramian for nonlinear stochastic systems in the light of Bayesian inference. First, define our Gramian, which refer to as estimability based on relation so-called Fisher Information Matrix initial state estimation. Then, study fundamental properties an empirical version Gramian. The practical usefulness proposed framework is examined through its application parameter and estimation natural gas engine cylinder.