Inferential motion control: Identification and robust control with unmeasured performance variables

Next-generation motion systems are expected to exhibit dominant flexible dynamical behavior. As a result, a dynamic relation between the measured variables and the performance variables is inevitable. The aim of the present paper is i) to develop a control framework to deal with unmeasured performance variables and ii) to implement the proposed methodology on a prototype experimental setup. A system identification for robust inferential control approach is pursued. Indeed, in the case that the performance variables cannot be measured directly, then these can be inferred from the measurements by means of a model. Experimental results i) confirm that prior approaches that aim at an improved response in terms of the measured variables can result in a deteriorated performance and ii) reveal that the proposed framework enables high performance robust inferential control.