Nonlinear continuous time generalized predictive controller for chaotic systems

The present study compares the performances of several types of continuous time generalized predictive controllers with nonlinear bang bang controller for controlling chaotic systems. Continuous time approach of controller design proposed in this study often exposes the fundamentals of underlying control problem which are obscured by sampling and therefore offers significant advantages over discrete time approach. The proposed controller works satisfactorily for both stabilization and tracking. By including proportional action into the standard nonlinear continuous time generalized predictive controller (NLCGPC) the speed and quality of response can be improved. Simulations on Lorenz and Rossler systems, operating in chaotic mode, demonstrate that NLCGPC can be a better alternative for controlling chaotic systems.