Real time nonlinear model predictive control induced by local linear controller

In this paper, a new nonlinear model predictive control (NMPC) algorithm guided by local linear controller is presented so that it is possible to be implemented in real time and obtain good performance. Firstly, in order to ensure the closed loop stability, a generalized pointwise min-norm (GPMN) controller formulation based on control Lyapunov functions (CLF) are given and parameterized. And then, the parameterized GPMN scheme is combined with the traditional NMPC algorithm. And the optimization variables in traditional NMPC are replaced by the parameters in GPMN, so that the computational burden of original NMPC algorithm can be reduced greatly. Finally, numerical simulations are conducted to verify the feasibility and validity of the new algorithms.