Robust receding horizon control based on data mining of nonlinear systems

In this paper, a data mining method is proposed to modeling the nonlinear system with constraints based on the sample data. The nonlinear system is expressed as a dynamic fuzzy system with modeling uncertainties. A robust receding horizon control scheme is proposed to stabilize the nonlinear system. The Min-Max optimal control problem arising in the robust receding horizon control is solved by the numerical method named finite difference with sigmoidal transformation according to dynamic programming principle. The obtained value function is used as a design parameter. This robust receding horizon controller can be applied in a real-time environment.