Aperiodic MPC for nonlinear systems with additive disturbances and input delays: A prediction self-triggered approach

This paper is concerned with the control problem of continuous-time nonlinear systems with additive disturbances and input delays. The main focus is to predict the actual state of the system at the triggering moment within the delay time caused by the controller and, on the other hand, to obtain the optimal input by solving the optimization problem based on this state advance delay time to compensate for the input delay. Based on the event-based strategy, a new prediction self-triggered model predictive control (PST-MPC) scheme is designed by considering the prediction error, delay time, and bounded disturbances, which ensures the desired control effect while decreasing the computational load. In addition, sufficient conditions for the stability of the closed-loop system and the recursive feasibility of the PST-MPC algorithm are developed, respectively. Finally, the validity of the presented method is proved by a simulation comparison. •A triggering mechanism considering the state prediction error and additive disturbances is derived.•A dual-mode PST-MPC algorithm is designed to reduce the frequency of solving OCPs.•Sufficient conditions of feasibility and closed-loop stability of the PST-MPC algorithm are obtained.•The minimum interaction time to avoid the Zeno phenomenon is given.