A robust finite-time convergent hybrid observer for linear systems

Motivated by the design of observers with good performance and robustness to measurement noise, the problem of estimating the state of a linear time-invariant system in finite time and with robustness to a class of measurement noise is considered. Using a hybrid systems framework, a hybrid observer producing an estimate that converges to the plant state in finite time, even under unknown constant (e.g., bias) and piecewise constant noise is presented. The stability and robustness properties of the observer are shown analytically and validated numerically.