A zone control strategy for stochastic model predictive control

The objective of this work is to extend the zone control approach from model predictive control to its stochastic counterpart. The motivation behind this idea is that zones are a natural extension of the concept of fixed set-point for systems subject to persistent disturbances. Therefore, it is possible to keep the outputs in desired regions without excessive control actions. We consider linear systems subject to bounded additive noise and propose a zone control formulation that includes hard constraints on the inputs and soft constraints on the states. The performance of the proposed approach is compared to zone controller that ignores disturbances and the results show a significant reduction of control moves, which means that the stochastic controller represents an interesting alternative to reduce the cost associated with control actions and to enhance the useful life of actuators.