last modified February 12, 2014 - 15:06 CET

Virtual sensors

Virtual sensors are estimators of product properties or process conditions using mathematical models rather than physical sensors.

Circuit implementation of virtual sensors offers low power consumption, fast response times and, at least for high volume applications, low cost and thus is highly desirable whenever a process state cannot be measured by a physical device, or a real sensor is too slow or too expensive to fabricate or to maintain.

We apply PWL functions to model identification in order to represent estimation laws, such as those used in virtual sensors to replace expensive physical sensors with estimates inferred from other measurements.

Involved People:

Main publications:

  • Rubagotti M, Poggi T, Oliveri A, Pascucci CA, Bemporad A and Storace M (2014), "Low-complexity piecewise-affine virtual sensors: theory and design", International Journal of Control. Vol. 87(3), pp. 622-632.
  • Poggi T, Rubagotti M, Bemporad A and Storace M (2012), "High-speed piecewise affine virtual sensors", IEEE Transactions on Industrial Electronics., Feb., 2012. Vol. 59(2), pp. 1228-1237.


  • This research activity was partially supported by the European Community through the MOBY-DIC project (FP7-IST-248858)

This site conforms to the following standards: