Model predictive control for max-plus-linear systems via optimistic optimization

The model predictive control problem for max-plus-linear discrete-event systems generally leads to a nonlinear optimization problem, which may be hard to solve efficiently. In this paper, we propose to apply optimistic optimization to resolve this problem. The algorithm builds a tree where each selected control sequence corresponds to a node of the tree. An optimistic exploration of the tree is implemented, where the most promising control sequences are explored first. We give an example to illustrate the effectiveness of the method.