Rolling self-triggered model predictive control for autonomous underwater vehicles with additional disturbances

To address the issue of stringent disturbance boundaries imposed by traditional self-triggered Model Predictive Control (MPC), which uses equal control and prediction horizons with a fixed sampling interval, this paper introduces a rolling self-triggered MPC strategy for trajectory tracking of Autonomous Underwater Vehicles. Firstly, by utilizing single-mode MPC with a prediction horizon longer than the control horizon and stringent constraints, we expand the disturbance upper bounds. We also derive sufficient conditions for the recursive feasibility of the online optimal control problem. Secondly, communication occurs only at specified triggering instants, determined by dynamic threshold crossings of the cost function error, resulting in substantial savings of communication resources. Additionally, an optimal control problem is formulated incorporating communication load in the cost function, aimed at achieving effective control performance while minimizing communication costs. We demonstrate that the strategy ensures the feasibility of the algorithm and the stability of the closed-loop system, provided that the controller parameters satisfy the specified conditions. Simulation studies also illustrate the efficacy of this methodology.