H_∞ Robust Predictive Control of Nonlinear Systems with Constraints

For a class of discrete-time nonlinear systems with constraints on inputs and states,this paper employs the method of finite dimensional parameterization and presents an H_∞ robust predictive control algorithm which is based on closed-loop optimization.The algorithm combines the receding horizon principle of predictive control with both differential game theory and nonlinear H_∞ control theory,the control variables are parameterized into polynomial control variables by finite dimensional parameterization,and transition equilibriums of the system to be controlled are introduced into the closed-loop optimization.Thus,the algorithm can not only deal with uncertain systems but also reduce the computational complexity of the on-line closed-loop optimization.Furthermore,we prove that the algorithm is feasible and robustly stable for the system subjected to bounded uncertainties.Finally,a numerical simulation is utilized to verify the effectiveness of the algorithm.