Extremum Seeking-based Iterative Learning Linear MPC
For control engineers dealing with uncertain linear systems, this work offers an online learning approach to adapt MPC models, though it is an incremental combination of existing methods.
The paper proposes an iterative multi-variable extremum seeking-based learning MPC algorithm for linear time-invariant models with parametric uncertainties, and demonstrates its effectiveness on a DC servo motor control example.
In this work we study the problem of adaptive MPC for linear time-invariant uncertain models. We assume linear models with parametric uncertainties, and propose an iterative multi-variable extremum seeking (MES)-based learning MPC algorithm to learn on-line the uncertain parameters and update the MPC model. We show the effectiveness of this algorithm on a DC servo motor control example.