Static Controller Switching Approach to Constraint Fulfillment for Time-varying References and Its Experimental Validations

This paper proposes a switching control strategy to fulfill pointwise-in-time state and/or control constraints for a time-varying reference, using more than one static controller. A concept of a maximal output admissible set can lead the condition to fulfill the pointwise-in-time constraints. Based on the condition plus a linear programming technique, we construct a switching rule, which on-line choices the most appropriate one of all designed static controllers to escape constraint violations. Theoretical key to compose this method is parallel shifting of the maximal output admissible set corresponding to the reference signal. The shifting can expand an applicable class of a constrained control problem that includes time-varying cases of the references. For validation of this method, we make an experimental control system of a dc-motor position servomechanism, in which the reference signal can be also inputted by a joystick to become time-varying, and demonstrate experimental validations to show this method's effectiveness under constant and time-varying reference signals respectively.