Event-Based MPC for Nonlinear Systems with Additive Disturbances: A Quasi-Differential Type Approach

Abstract

By integrating the event-based mechanism and the model predictive control (MPC) method, an improved event-based MPC framework is constructed for the nonlinear control problem subject to disturbances. Firstly, a new event-triggering condition is suggested, which is constructed on the basis of the gradients of the differences between the optimal state prediction and the actual one at two consecutive sample times. Then, an event-based MPC algorithm is further proposed, in which the dual-mode control technique is incorporated to handle the nonlinear perturbed system. Furthermore, it is strictly demonstrated that the proposed algorithm will ensure the feasibility of the MPC method and the stability of the considered system, while significantly decreasing the number of solving optimization problems, based on which resources for information transfer can be effectively saved. Finally, simulations and comparisons are shown to verify the efficacy of the proposed framework.