NSGA-II based high gain observer improved optimization method

This paper purpose is to optimize a high gain state observer for the estimation nonlinear chaotic systems.The upgrading method, based on the use of a non dominated sorting genetic algorithm (NSGA-II), relies on a quadratic optimization fitness function is presented to generate the most suitable value of the observer influential parameter θ that define the observation gain. NSGA-II algorithm is considered as a competent multiobjective exploration approach. In fact, the proposed criteria grants an adjustment of the observation error taking into consideration the correction factor of the observer. Furthermore, a remarkable specification of the proposed optimization approach is its independence to initial conditions allowing to override the problem of suboptimal conditions, which are widely used in optimization methods. Experimental simulation is proposed to illustrate the efficiency and prominent results of the designed observation approach, applied to state reconstruction of the well-known unified nonlinear perturbed chaotic systems.