Set-theoretic Methods in Robust Detection and Isolation of Sensor Faults


F. Xu, Puig, V., Ocampo-Martinez, C., Olaru, S., and Stoican, F., “Set-theoretic Methods in Robust Detection and Isolation of Sensor Faults”, International Journal of Systems Science, vol. 46, no. 13, p. 18, 2015.

Start Page:



In this paper, we consider a multi-sensor networked control configuration with linear plant which is affected by a bounded additive disturbance. Shared network is used for the communication between sensors and controller. It is assumed that the sensors are prone to abrupt faults, while the controller’s input may be updated with a varying time-delay. In order to identify and isolate the sensor(s) providing faulty information, we equip the controller with a set-based detection and isolation routine. Furthermore, in the case when the network induces time-delays, control is performed based on the knowledge we have on the mathematical model of the plant. In the presence of model inaccuracies or disturbance, such a control action may not guarantee satisfying performance of the system. Therefore, a stabilising controller with delay compensation has been designed. The functioning of the proposed control algorithm has been illustrated through an example.