Virtual Sensors and Actuators

Abstract

This chapter introduces the design of virtual sensors and actuators using the classical eigenvalue assignment approach, widely used for the design of controllers and observers in state-space. It presents a linear matrix inequality (LMI)-based procedure. The attractiveness of this solution is that the virtual sensor/actuator technique described so far can be extended easily to work with the nonlinear systems described by convex representations, such as the linear parameter varying system. The chapter illustrates the virtual sensor and actuator approach using a well-known case study: the four-tank system. It also presents a separate formulation of virtual sensors/virtual actuators. The main advantage of the LMI-based design is the fact that it enables the extension of the virtual sensor/actuator technique to linear parameter varying systems. The chapter concludes with a presentation of the conclusions and some outlooks on the current trends of virtual sensors and actuators.