Global adaptive synchronization based upon polytopic observers

A new adaptive synchronization scheme achieving a joint state and parameter estimation is proposed as an alternative to the well-known Extended Kalman Filter based on a first order approximation. It involves a polytopic observer which no longer requires a linearization. The resulting adaptive synchronization is global and holds for a large class of chaotic systems. The computation of the gain of the observer is performed in a systematic way and is derived from the solutions of a Linear Matrix Inequalities set.