Stochastic adaptive tracker based on noise-corrupted space-time measurement process

A multiple model adaptive estimator (MMAE) is formulated to estimate the state of a dynamic system modeled by a linear stochastic differential equation, from which the feedback observations are described as a noise-corrupted Poisson space-time point process. Then the MMAE is embedded into a stochastic adaptive PI (proportional-plus-integral) tracker using LQG (linear quadratic Gaussian) and assumed certainty equivalence techniques. The MMAE, the Kalman filter (used to estimate the target state), and the tracker are evaluated for parameter sensitivity, robustness, and adaptation using Monte Carlo simulation.<
>